Printer Friendly

Net exchange of greenhouse gases from soils in an unimproved pasture and regenerating indigenous Kunzea ericoides shrubland in New Zealand.

Introduction

The conversion of forest to pasture for grazing animals became widespread in New Zealand from the mid 19th century (Harding 2003). Indigenous shrubland species Kunzea ericoides var. ericoides (A. Rich) J. Thompson (kanuka) and Leptospermum scoparium J. R. Forst & G. Forst. (manuka) were restricted to inaccessible, steep, or upland sites not suited to agricultural use. However, with decline in the economy of pastoral hill country farming in the last 2 decades, the abandonment of grazing has led to the natural reversion of pasture back to shrubland. In areas where seed sources are available, Kunzea has re-established and thrived (Wilson 1987) and presently represents 10% of New Zealand's total land area (Tate et al. 2005).

Kunzea and Leptospermum are primary successional species and occur widely across New Zealand (Allen et al. 2009). Invasion of pasture by Kunzea is slow because of competition for soil resources and available light (Reay and Norton 1999), but once established, it can show high rates of biomass growth and survive for up to 80 years (Trotter et al. 2005).

More recently, natural regeneration of indigenous shrubland on economically marginal land has been promoted for pastoral farming in New Zealand for erosion control and as a way of establishing new carbon sinks to offset greenhouse gas emissions. Trotter et al. (2005) estimated that there is potential for 1.5 Mha of steep, marginal hill country with seed sources available that could sequester carbon in shrubland biomass at rates of 1.9-2.5 MgC/ha.year over a period of 40 years regrowth. This would make a substantial contribution to offsetting New Zealand's national net greenhouse gas emissions. Depending on the price of carbon on trading markets, such 'carbon farming' using regenerated shrubland could become an attractive alternative land use for land owners.

When considering the potential for regenerating shrubland to offset greenhouse gas emissions through increased C stored in vegetation biomass, it is also important to consider changes in sources and sinks of carbon dioxide (C[O.sub.2]), methane (C[H.sub.4]), and nitrous oxide ([N.sub.2]O) from the soil. Globally, more than two-thirds of the carbon stored in terrestrial systems is stored in the soil (Amundson 2001) and up to 80% of respiration from forest ecosystems can be attributed to soil C[O.sub.2] emissions (Curtis et al. 2005). Rates of C[O.sub.2] emissions increase strongly with increasing soil temperature (Lloyd and Taylor 1994) and decrease as root-zone water content falls below field capacity (Brown et al. 2008).

Reversion of pasture to shrubland can lead to enhanced oxidation of C[H.sub.4] due to methanotrophic activity that is particularly pronounced in forest soils (Smith et al. 2000; Price et al. 2004; Tate et al. 2007). This can enhance the opportunity to reduce net greenhouse gas emissions with land-use change. In Europe, a study by Prieme et al. (1997) showed that the reversion of woodland on cultivated arable soils led to slowly increasing C[H.sub.4] oxidation rates that took at least 100 years to reach those measured in undisturbed woodland soils. Decreases in the rates of C[H.sub.4] oxidation have been measured in soils in tropical (Steudler et al. 1996; Verchot et al. 2000) and temperate (Tate et al. 2006) regions following forest clearance. Although soil is recognised as a natural sink for C[H.sub.4] (Lowe 2006), data are generally lacking for changes in this sink with land-use change, especially for regenerating shrublands. Rates of C[H.sub.4] oxidation are strongly dependent on root-zone water content (Price et al. 2004; Tate et al. 2007) and soil physical properties (Del Grosso et al. 2000), with only weak dependence on soil temperature (Price et al. 2004).

Very little attention has been given to measuring changes in net [N.sub.2]O emissions from soil with land-use change from pasture to shrubland. While it is known that increased nitrification and increased [N.sub.2]O emissions can occur shortly after clear felling mature trees (Zerva and Mencuccini 2005), a study in New Zealand showed that clear-felling Pinus radiata D. Don did not result overall in enhanced emissions, but did cause a sharp spike in emissions immediately after harvest (Tate et al. 2006).

The objectives of this study were to quantify the net greenhouse gas exchange of C[O.sub.2], C[H.sub.4], and [N.sub.2]O from soils of regenerating shrubland and any changes to the rates of gas exchange with increasing age of shrubland. We achieved this by making measurements at 3 sites along a chronosequence consisting of unimproved pasture (UP), young (8-12 years) Kunzea trees (YK), and old (80 years) Kunzea trees (OK). Measurements of the rates of exchange of C[O.sub.2], C[H.sub.4], and [N.sub.2]O were made monthly with concomitant measurements of root-zone water content and soil temperature to determine the responses to these driving variables. These responses were then used in models with daily values of root-zone water content and soil temperature for 1 year to estimate annual rates of exchange of the 3 greenhouse gases from the soil at the 3 sites.

Materials and methods

Site description

The 3 sites were located at Little River, 70 km south-east of Christchurch, New Zealand (43.74[degrees]S, 172.85[degrees]E, elevation above sea level 180 m). The climate at the site is maritime with the long-term average rainfall being I 115 mm. Summer conditions are warm (mean daily maximum temperature in January and February is 22.2[degrees]C) (Anon. 1983).

The 3 sites represent a successional sequence typical of a regenerating Kunzea ericoides var. ericoides (A. Rich) J. Thompson (kanuka) shrubland. At each site, an area ~10 by 10 m was used for measurements. The UP site was dominated by brown top (Agrostis capillaries L.) and creeping bent (Agrostis stolonifera L.) with Kunzea absent. The YK site consisted of trees 8-12 years old with height ranging from 1 to 3 m. The trees at the OK site were -80 years old with a height range of 4-5 m. The UP and YK sites were grazed before this study, but animals were excluded once the study commenced to allow a focus on the impacts of removing animals on the shrubland regeneration and associated greenhouse gas fluxes. The OK site had been undisturbed by animals for at least 6 years before the start of the study.

The soil at the location was a Pawson silt loam (colluvium variant) (Trangmar 1986) consisting of imperfectly drained loess derived from greywacke with evidence of a clay-enriched B horizon (Hewitt 1998). More specifically, an Immature Pallic soil was present at the UP and YK sites and a Fragic Pallic soil at the OK site, following the New Zealand Soil Classification (Hewitt 1998), or a Aquic Haplustept following the USDA Soil Classification (T. Webb, Landcare Research, pers. comm., 2008). The mean ([+ or -] standard error) litter layer depth at the OK site measured at 6 locations was 13 [+ or -] 2.5 mm. The range in total soil carbon content was 4.80--1.92% at depths of 0-100 and 100-200 mm, respectively, and total soil nitrogen content ranged from 0.33 to 0.15% at the same depths, respectively.

Soil physical properties (Table 1) were measured following Klute (1986). For total carbon and nitrogen analyses, 4 replicate soil cores were taken, bulked, air-dried, and ground to <2 mm for analysis using an elemental analyser (Model CNS-2000, LECO Australia Pty Ltd). Soil mineral nitrogen (N[H.sub.4.sup.+] and N[O.sub.3.sup.- ]) was extracted on samples collected seasonally, as [N.sub.2]O fluxes from the regenerating shrubland in the absence of nitrogen inputs were likely to be very low. Extractions were done using 2 M KCl and analysed using a flow injection analyser (Model FIAstar 5000, Foss Tecator AB, Hoganas, Sweden) following Blakemore et al. (1987). Soil pH (d[H.sub.2]O) was measured on bulked samples using 3 replicates from the three sites for depths of 0-50, 50-100, and 100-150 mm using a 1 : 2.5 soil : solution ratio.

Meteorological and soil measurements

A weather station was established in a clearing in the YK/UP site and measurements of rainfall, soil temperature, and root-zone water content (Model ML1 ThetaProbe, Delta-T, UK) were made continuously. The data were recorded using dataloggers and stored as half-hourly averages. Measurements of root-zone water content, [theta], were made at both the YK and OK sites, and because of the close proximity (5 m) of the UP and YK sites, similarity in the measurements for these sites was assumed. Independent monthly data from integrated TDR measurements (over a depth of 300 mm) for the duration of the study supported this assumption with the relationship: [[theta].sub.UP] = 1.025 YK [[theta].sub.YK] + 0.049, [r.sup.2] = 0.88. One soil water content probe at the YK and OK sites was positioned at 50, 200, and 500 mm. The probes were calibrated using episodic gravimetric analysis on soils taken from near the sensors.

Gas sampling procedures

Samples of C[O.sub.2], C[H.sub.4], and [N.sub.2]O were collected at monthly intervals (Price et al. 2004; Fang et al. 2009; Ussiri et al. 2009) from September 2005 to September 2006 from closed chambers placed on tings set permanently into the soil surface to a depth of 50 mm. The chambers used at the OK and YK sites for sampling C[H.sub.4] and [N.sub.2]O were transparent polyethylene domes (170 mm diameter, 130 mm height), and at the UP site, the chambers were cylindrical and constructed of PVC (240 mm diameter, height 30 mm). Different-sized chambers were used to accommodate differences in the magnitude of fluxes, especially C[H.sub.4]. The chambers were clamped on to the rings fitted with a gas-tight foam seal for each measurement (40 min duration), with 3 samples being taken from each chamber at 0, 20, and 40-min intervals. Gas samples (10 mL) were taken using greased glass 'Van Mark' interchangeable syringes and injected into pre-evacuated 6.5-mL septum-capped glass vials (Labco Exetainer, High Wycombe, UK) following Price et al. (2004). Samples from 6 replicate chambers from each site were transported to the laboratory and brought to atmospheric pressure immediately before analysis using automated gas chromatography (Model SRI8610C, Torrance, California, USA) with a flame ionisation detector and an electron capture detector for analysing C[H.sub.4] and [N.sub.2]O, respectively (Clough et al. 1998). Fluxes of C[H.sub.4] and [N.sub.2]O were calculated from the change in concentration over a set period and the surface area of each ring following Price et al. (2004).

Measurements of C[O.sub.2] exchange were made directly using a soil respiration chamber (Models EGM3 and SRC1, PP Systems, UK) that was placed onto tings (12 mm diameter, 50 mm depth) set permanently into the soil surface to a depth of 25 mm. At each site there were 16 replicate rings, and at the YK and UP sites, the grass was clipped to ground level before measurements were made. Soil temperature was measured at the same time at a depth of 50 mm using a thermocouple.

Data analyses

All fluxes for C[O.sub.2], C[H.sub.4], and [N.sub.2]O are presented as means ([+ or -] standard error) of the measurements at each site and the results were considered to be significantly different if P < 0.05.

A non-linear mixed effect model (Pinheiro and Bates 2000) was used to describe the response of fluxes of C[O.sub.2] to soil temperature, [T.sub.s] (K), and root-zone water content, [theta], of the form following Lloyd and Taylor (1994) and Brown et al. (2008) as follows:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (1)

where [R.sub.s] is the C[O.sub.2] flux, [R.sub.10] is the base value of [R.sub.s] at a soil temperature of 10[degrees]C, [E.sub.o] is a parameter related to the energy of activation, and [theta]c is the value of [theta] above which [R.sub.s] remains constant.

For C[H.sub.4] and [N.sub.2]O, a linear mixed effects model was used to describe the seasonal response of fluxes to [T.sub.s] and [theta] of the form:

f = a + [b.sub.1][T.sub.s] + [b.sub.2][theta]+ [b.sub.3]([T.sub.s][theta]) (2)

where f is the flux for C[H.sub.4] or [N.sub.2]O, and a and b are parameters. For [N.sub.2]O the data were transformed using an arbitrary power function of the mean to account for heteroscedascity and to normalise the residuals using standard statistical procedures (R, v2.6.2, R Development Core Team 2008).

Results

Environmental and soil variables

Mean daily air temperature varied between 4.9[degrees]C in winter (June) and 31.1[degrees]C in summer (January) and [T.sub.s] at 50 mm depth closely followed the same trends (Fig. 1). The maximum average daily [T.sub.s] (at 50 mm depth) recorded during the measurement period was 20.4[degrees]C at the UP site in summer (January). While ground frosts did occur during winter, they were not prolonged and were unlikely to impact on [N.sub.2]O emissions.

[FIGURE 1 OMITTED]

Rainfall at the UP and YK sites during the year of measurement was 1179 mm, with larger amounts falling in autumn and winter than in the other seasons. A continuous period without rain for 24 days in spring starting on 11 October resulted in a marked decrease in [theta] (Fig. 1). This remained low until the following autumn when a large rainstorm of 111.2 mm fell on 12 May 2006. Variability in [theta] at 50 mm depth was greater at the OK site than at the other sites, possibly due to the presence of a litter layer with a high capacity to retain water. For nearly 8 months, values of [theta] at all sites were below field capacity (Table 1) and lower at the OK site than the others, consistent with high rainfall interception by the closed tree canopy.

Measurements of C[O.sub.2] fluxes

There were marked differences in C[O.sub.2] flux both seasonally and between sites (Fig. 2a). Seasonal changes were most marked at the UP site. For individual months, the highest (0.67 [+ or -] 0.05 g C[O.sub.2]/[m.sup.2].h [+ or -] 1 s.e.) and lowest (0.09 [+ or -] 0.01 g C[O.sub.2]/[m.sup.2].h [+ or -] 1 s.e.) fluxes were measured at the UP site in spring (October) and early winter (June), respectively. During spring and summer months (November-February), the highest fluxes were measured at the UP site. In contrast, in autumn and winter (April October) the highest values were recorded at the OK site. Carbon dioxide fluxes at the YK site fell between the values for the UP and YK sites for every month.

[FIGURE 2 OMITTED]

Measurements of C[H.sub.4] fluxes

Mean ([+ or -] s.e) monthly measured fluxes of C[H.sub.4] were -9.6 [+ or -] 2.4, -7.5 [+ or -] 6.4, and 54.0 [+ or -] 6.6 [micro]g C[H.sub.4]/[m.sup.2].h at the UP, YK, and OK sites, respectively (Fig. 2b). The negative sign denotes that, when averaged over the year, the sites were all net sinks for C[H.sub.4]. Sporadic net C[H.sub.4] emissions occurred at all sites across the measurement period but they were most pronounced at the YK site. In June, July, and September 2006 at the UP site and August and September at the YK site, C[H.sub.4] fluxes were below the minimum detectable value of 2.7 and 9.8 [micro]g C[H.sub.4]/[m.sup.2].h for the UP and YK/OK sites, respectively, as determined from a series of repeated analyses of ambient air (van der Weerden 1999, unpubl, obs.) and thus were considered to be zero. There was a strong seasonal dependence of C[H.sub.4] fluxes at the YK and OK sites but this was not apparent at the UP site. The seasonal variability at the YK site was much larger than the variability at the UP site. The maximum rate of oxidation (-90.4 [micro]g C[H.sub.4]/[m.sup.2].h) was measured at the OK site in late summer (March) and it was similar to the average C[H.sub.4] oxidation rate measured in an old growth beech forest (Price et al. 2004).

Measurements of [N.sub.2]O fluxes

Mean [N.sub.2]O flux densities were low at all sites over the entire measurement period (Fig. 2c) with a maximum of 7.9 [micro]g [N.sub.2]O/ [m.sup.2].h in late autumn (May) at the OK site. Fluxes exceeded the minimum detectable value (1.60 [micro]g [N.sub.2]O/[m.sup.2].h) only at the OK site. Fluxes were low during the summer months (January-March) and in late winter (June-August). Variability in the measurements in late autumn (April-June) was high.

Modelling net greenhouse gas exchange

Seasonal variability in the measurements of C[O.sub.2] flux at 2 of the 3 sites (UP and YK) was simulated well by the model with [T.sub.s] and [theta] as driving variables (Eqn 1) ([r.sup.2] values for the UP and YK sites were 0.84 and 0.54, respectively). Values for all the parameters at all sites were highly significant (Table 2).

Values for [R.sub.10] at the UP and YK sites were similar but the value at the OK site was much higher. Values of [[theta].sub.c] were similar at all sites, suggesting that C[O.sub.2] flux was not limited at the sites until [theta] fell below 0.27-0.29 [m.sup.3]/[m.sup.3]. This is lower than the water content at field capacity, but much higher than the value at permanent wilting point (Table 1).

Fitting the model in Eqn 2 to the C[H.sub.4] flux data showed that the dominant driving variable varied with vegetation type. At the UP site, when the effects of both [theta] and [T.sub.s] were analysed together, [T.sub.s] was the dominant variable (Table 3). However, when [T.sub.s] was removed from the analysis, the response to [theta] was highly significant. At the OK site, C[H.sub.4] flux was related significantly to [theta] but there were no significant relationships with [T.sub.s] or [theta] at the YK site. Data at the OK (Fig. 3) and UP sites showed clear linear increases in C[H.sub.4] oxidation rate with decreasing [theta]. Data from September 2005 were excluded from the analysis because of spurious emissions from 1 replicate.

It was only possible to fit the model to the data for [N.sub.2]O at the OK site since the fluxes at the other sites were lower than the minimum detectable limits. The results showed that [N.sub.2]O fluxes were regulated by [T.sub.s] and [theta] and that there was also a significant interactive effect (P=0.01, Table 3).

Annual estimates of net greenhouse gas exchange

Annual total exchange of the 3 greenhouse gases at the 3 sites was estimated from values of the parameters obtained by fitting the data to the models (Eqns 1 and 2, Tables 2 and 3), combined with daily measurements of [T.sub.s] and [theta]. For the year starting in September 2005, annual C[O.sub.2] flux was highest at the OK site, almost double the value at the UP site (Table 4).

The annual C[O.sub.2] fluxes were very similar at the UP and YK sites. Estimates of net C[H.sub.4] exchange showed that both the UP and the OK sites were sinks for C[H.sub.4], but the OK site was much stronger. It was not possible to use the model at the YK site, so measured values from this site for each month were summed for the year to allow comparison with other sites. This suggested that the YK site was also a weak sink for C[H.sub.4], although this estimate did not take account of changes in root-zone temperature and water content during each month. Nitrous oxide emissions at the OK site were low, but were equivalent to 61% of the C[H.sub.4] oxidation rate at this site on a C[O.sub.2]-equivalent basis.

Discussion

Measurements on soils at the 3 sites along the chronosequence of pasture reversion to shrubland have provided a comparison of the net emissions of the principal greenhouse gases with land-use change. We acknowledge that our experimental design of 1 plot for each land use led to pseudo-replication in our statistical analysis, but this was dictated by difficulties in locating sites with adjacent land uses and restrictions on the numbers of intensive measurements that we were able to make. While our data can be used to demonstrate differences between plots, care is required when using them for quantitative spatial scaling across landscapes.

[FIGURE 3 OMITTED]

Greenhouse gas emissions were dominated by C[O.sub.2] release from the soil surface and this was much higher annually at the OK site than at the YK and UP sites (Table 3). Methane oxidation was also pronounced at the OK site where the annual total exceeded [N.sub.2]O emissions when expressed on a C[O.sub.2]-equivalent basis. The same was true at the UP site, but the net exchange of both gases was much lower. Although the annual estimate of C[H.sub.4] oxidation at the OK site (127.3 kg C[O.sub.2]-e/ha.year) was small compared with estimated annual average C[O.sub.2] uptake rates for shrubland (-8.1 x [10.sup.3] kg C[O.sub.2]/ ha.year; Trotter et al. 2005), our data confirm that soil oxidation of C[H.sub.4] in shrubland systems could be considered as a potential mitigation option for reducing net emissions.

The Arrhenius model used to describe the response of C[O.sub.2] flux to temperature has been well tested (Lloyd and Taylor 1994) and, when combined with a response to [theta], is used widely to explain the variability observed in field measurements (Davidson et al. 1998; Kelliher et al. 1999; Hunt et al. 2002; Brown et al. 2008). A recent re-evaluation of the processes regulating the relationship between C[O.sub.2] flux and [theta] by Cook and Orchard (2008) confirmed the suitability of a linear relationship to describe this response. The values for [R.sub.10] at the OK site were almost twice that at the UP site, reflecting differences associated with land use (Table 2). Although high root length density in managed pasture would be expected to result in a high value for [R.sub.10], the unimproved pasture at our site had been abandoned with no applications of fertiliser for many years before our measurements. Higher [R.sub.10] of 0.4 and 0.76 g C[O.sub.2]/[m.sup.2].h for a managed grazed pasture growing on a drained peat (Nieveen et al. 2005) and a well-irrigated productive grazed pasture (Brown et al. 2008) were most likely attributable to high root metabolic activity. Measurements of C[O.sub.2] flux were, however, low at other sites in New Zealand where low productivity was attributable to limited water availability. Arneth et al. (1998) measured a range in C[O.sub.2] fluxes from 0.16 to >0.63 g C[O.sub.2]/[m.sup.2].h for soils under a dryland Pinus radiata forest and Hunt et al. (2002) reported similarly low C[O.sub.2] fluxes of 0.02-0.27 g C[O.sub.2]/ [m.sup.2].h in summer at a dryland tussock site. However, at the tussock site, C[O.sub.2] flux increased to 0.43 g C[O.sub.2]/[m.sup.2].h when water was added to the soil to simulate the effects of rain. Ross et al. (2002) found that C[O.sub.2] fluxes were lower under Pinus radiata (12-30 years old) when compared with corresponding pasture sites. In a young regenerating stand of Pinus ponderosa in Oregon, USA, mean soil C[O.sub.2] flux was 0.32 g C[O.sub.2]/[m.sup.2].h (Law et al. 2001). The higher value of [R.sub.10] than those reported elsewhere at our OK site most likely resulted from high fine-root length density observed near the mineral soil surface and in the litter layer. The importance of root respiration was shown by Li et al. (2004) who reported a 56% reduction in soil respiration after 7 years of root exclusion in a tropical secondary forest ecosystem containing abundant shrubs. In a recent experiment to partition the sources of soil respiration in a mature Kunzea ericoides forest, Millard et al. (2010) demonstrated the contribution of roots to soil surface respiration by reporting that the C[O.sub.2] flux could be equally apportioned to roots and soil organic matter, demonstrating the contribution of roots to soil surface C[O.sub.2] emissions. Li et al. (2004) also demonstrated the importance of new litter to soil respiration showing it was responsible for 15% of the total soil C[O.sub.2] flux.

The values of the threshold parameter for [theta] below which soil respiration decreased, [theta]c, were similar at each of our sites, and showed that the C[O.sub.2] flux was limited by root-zone water content only when [theta] fell well below field capacity (Table 1). This low value of [theta] was reached throughout the measurement period from late spring (November) to late autumn (May), particularly at the OK site (Fig. 1). This sugges[T.sub.s], in contrast to the findings of Brown et al. (2008), that water availability limited C[O.sub.2] flux for much of the year when soil temperatures were above ~10[degrees]C. The study by Brown et al. (2008) was conducted near our site but in a grazed pasture growing on a heavy clay soil at a poorly drained site where soil surface C[O.sub.2] flux was rarely limited by [theta]. Similar limitations to C[O.sub.2] flux to those at our site have been reported by Risch and Frank (2006), where values down to 10% of the potential rate were observed in late summer in a temperate natural grassland. Apart from root activity, the observed variability in the C[O.sub.2] flux may also be due to changes in [T.sub.s] and [theta] that regulate the dynamic pool of dissolved organic carbon available for soil microbial activity (Sato and Seto 1999). It is also likely that dissolved organic carbon is an important substrate for C[H.sub.4] production and contributes to [N.sub.2]O production in the presence of a source of mineral nitrogen.

Methane oxidation was highest at the OK site (~80 years) and was more than 3 times the value at the UP site. Methanotrophic bacteria are most active within a zone close to the soil surface (Roslev et al. 1997) where rates of oxidation depend strongly on the availability of C[H.sub.4] as a substrate and [theta] that regulates oxygen diffusion and nitrogen availability (Conrad 1996). Other studies (Castro et al. 1994; Price et al. 2004) have shown a strong linear response of increasing C[H.sub.4] flux to decreasing soil [theta]. Such a linear response generally holds true for poorly draining silt loam soils found in agricultural situations (Li and Kelliher 2007) as well as for silt loam forest soils with lower bulk density (Price et al. 2004). Another study with lighter, sandy shrubland soils showed an exponential response (Castaldi and Fierro 2005), and a third study at low values of soil water content had a curvilinear response demonstrating an optimum soil [theta] for oxidation (Borken et al. 2006). However, the range in water content for our data was limited and this did not allow us to confirm the exact nature or form of the response, so for simplicity, we adopted a linear relationship within the range of soil water contents at our sites.

Higher rates of C[H.sub.4] oxidation at the YK and OK sites compared with those at the UP site were most likely attributable to lower [theta] during the spring and summer periods (Fig. 1) resulting from increasing rates of rainfall interception by the developing canopies (Rowe et al. 1999), leading to more favourable conditions for the increase in populations of methanotrophic bacteria and their activity. Our results confirm earlier work (Price et al. 2004) showing the sensitivity of C[H.sub.4] oxidation rates to [theta] (Fig. 3) with little influence of [T.sub.s].

The rates of C[H.sub.4] oxidation at the UP site are similar to values measured for grazed pasture elsewhere (Judd et al. 1999; van den Pol-van Dasselaar et al. 1999; Li and Kelliher 2007; Saggar et al. 2008). However, C[H.sub.4] oxidation rates at the UP site varied more widely showing similar emission but much greater oxidation rates when compared with an improved clover-grass pasture in south-west Australia with a drier climate (760 mm) (Livesley et al. 2009). Additional work involving more frequent measurements is required to explain the variability in greenhouse gas fluxes, especially C[H.sub.4], in developing ecosystems. Likewise, our rates measured at the YK site are similar to those reported for other re-afforested areas elsewhere (Suwanwaree and Robertson 2005; Tate et al. 2007). However, C[H.sub.4] oxidation rates in an 8.5-year-old blue gum (Eucalyptus globulus) plantation in a Mediterranean climate varied over a much smaller range (than those measured in the YK) with the maximum flux being close to -15 [micro]g C[H.sub.4]/ [m.sup.2] x h (Allen et al. 2009). At the YK site, the maximum C[H.sub.4] oxidation rate of -49.7 [micro]g C[H.sub.4] [m.sub.2] x h was measured in December 2005, suggesting that the methanotrophic populations were sufficiently large and active to provide oxidation rates close to half the maximum oxidation rate at the OK site (-90.4 [micro]g C[H.sub.4]/[m.sup.2] x h) when [theta] was low (0.22 [m.sup.3]/[m.sup.3]) and [T.sub.s] moderate (15.4[degrees]C). The sporadic C[H.sub.4] emissions measured in YK may have been due to the presence of some damper anaerobic hotspots of C[H.sub.4] production within the soil, which, when combined with of patches of more readily degradable litter, may have acted as a substrate for C[H.sub.4] production. Further, the additional moisture and litter may have attracted more macro-fauna which have in themselves been shown to contribute to C[H.sub.4] production in localised areas in the soil (Kammann et al. 2009).

Oxidation rates at the OK site are higher than the value of -.25 kg C[H.sub.4]/ha x year reported by Dutaur and Verchot (2007) for a chaparral community growing in a Mediterranean climate but 50% lower than those measured in an undisturbed Nothofagus forest soil (-10.5 kg C[H.sub.4]/ha x year) by Price et al. (2004). This difference is most likely due to the relatively young age of the trees at the OK site. Rates of C[H.sub.4] oxidation are known to increase in relation to forest successional status as soil physical properties and methanotroph populations change (Tate et al. 2007). Suwanwaree and Robertson (2005) measured rates of oxidation that were 10% greater in a 10-year-old woody ecosystem than they were in an arable soil. A much larger increase of 75% was observed at the same site in plots where trees had regenerated 40-60 years after agriculture ceased. Likewise, Tate et al. (2007) measured a 50% increase in C[H.sub.4] oxidation rates in a 50-year-old Kunzea shrubland compared with a nearby pasture soil. Where physical disturbance has been a feature of managing agricultural soils, considerable time may be needed before a viable methanotroph population can establish and C[H.sub.4] oxidation rates increase. For example, Maljanen et al. (2001) measured C[H.sub.4] emissions for 23 years after afforestation of boreal agricultural peatland soils and Prieme et al. (1997) reported that 100-200 years was needed before C[H.sub.4] oxidation reached maximum rates as woodland recovered on sites used formerly for agriculture.

Nitrous oxide emissions from our 3 sites were low compared with values from agricultural sites (Luo et al. 2008; Saggar et al. 2008) and other temperate forest soils (Morishita et al. 2007). The highest fluxes were measured at the OK site, largely due to higher soil N[O.sub.3.sup.-] nitrogen throughout the profile to a depth of 150 mm in particular (Table 1), consistent with the findings of Wallenstein et al. (2006) and Dalai and Allen (2008). The measured values of [N.sub.2]0 flux at the UP site were below the minimum detectable limit, consistent with the low values reported for all [N.sub.2]O sources (including denitrification) for an unfertilised agricultural soil in New Zealand (Parfitt et al. 2006). These low rates can be explained by the tight cycling of nitrogen in indigenous ecosystems (Ross et al. 1996). Mineral-N needs to be included as a component in the model describing the [N.sub.2]O fluxes in future studies, but it is unlikely to have impacted on the results at our site as the [N.sub.2]O fluxes were so low. The lower fluxes measured in June and July may be due to increased denitrification, possibly leading to more [N.sub.2] rather than [N.sub.2]O production. It is also possible that the low phosphate retention or anion exchange capacity of the soils (Trangmar 1986) led to the low availability of nitrate necessary for denitrification. However, nitrate retention in organic matter at the OK site was higher than that at the other sites, so this could have contributed to higher [N.sub.2]O emissions. There are few other reports of measurements of [N.sub.2]O fluxes for shrubland soils. At an arid shrub-steppe site in south-central Washington (USA), [N.sub.2]O emissions were <0.1 kg [N.sub.2]O-N/ha x year with a maximum of 2.7 kg [N.sub.2]O-N/ha x year following a rainfall event (Mummey et al. 1997). Nitrous oxide emissions from temperate indigenous forests and Eucalyptus plantations in Australia (Dalal and Allen 2008) were similar to those measured at our site and the mean [N.sub.2]O flux for Japanese temperate forest soils was 0.26 kg [N.sub.2]O/ha x year (Morishita et al. 2007).

Forest regeneration may result in increased [N.sub.2]O emissions (Ball et al. 2002, 2007) and emissions from mature stands can be greater than those in adjacent, old grazed pastures (Erickson et al. 2001). This is attributable to changes in the ratio of soil nitrogen to phosphorus (Davidson et al. 2007). Removal of trees results in a flush of [N.sub.2]O as the forest floor is disturbed and nitrogen is mineralised (Steudler et al. 1991; Tate et al. 2006). Because of the high global warming potential of [N.sub.2]O, land-use change can result in significant changes in net emissions due to increased emissions of this potent greenhouse gas. While our data confirm low rates of [N.sub.2]O emissions at all 3 sites, caution is needed when scaling to larger areas because of differences in nitrogen cycling and availability in relation to landscape and environmental conditions.

In conclusion, we have used measurements and models to explain seasonal variability in measurements of C[O.sub.2], C[H.sub.4], and [N.sub.2]O fluxes at 3 sites along a sequence of naturally regenerating shrubland. Net soil greenhouse gas exchange at the 3 sites was dominated by C[O.sub.2] flux densities and these were regulated seasonally by variability in both [T.sub.s] and [theta]. Methane oxidation rates on average were the highest at the OK site (~80 years old) and were more than 5 times that measured at the UP site. At the YK site the maximum C[H.sub.4] oxidation rate reached 54% of that at the OK site in the early summer (when [T.sub.s] was moderate and [theta] was low), indicating that there is potential for rapid recovery of methanotropic populations when comparing maximum C[H.sub.4] fluxes over a short time frame after the establishment of shrubland on pasture. However, on an annual basis our data suggest that C[H.sub.4] oxidation rates decrease as land reverts from unimproved pasture to shrubland. Nationally, promotion of increasing natural regeneration of shrublands and forests (Trotter et al. 2005), and more prolonged drought predicted for eastern regions of New Zealand with changing climatic conditions, are likely to result in enhanced potential for C[H.sub.4] oxidation, thereby reducing net national greenhouse gas (C[H.sub.4]) emissions (relative to 1990 baseline values for C[H.sub.4]). More research is thus required to enhance the efficacy of the natural process of soil C[H.sub.4] oxidation so that it can become a realistic C[H.sub.4] mitigation tool. Work should be also be undertaken on utilising the ability of the methanotrophs to rapidly recover C[H.sub.4] oxidising activity under naturally regenerating ecosystems.

Nitrous oxide emissions at the 3 sites were low and could be measured with confidence only at the OK site. Annual [N.sub.2]O emissions were more than offset in C[O.sub.2]-equivalent terms by C[H.sub.4] oxidation at the UP and OK sites. Our results demonstrate that changes in net emissions can be anticipated with land-use change, and that accounting for all 3 major greenhouse gases provides a more robust assessment of greenhouse gas inventories at large spatial scales.

10.1071/SR09156

Acknowledgments

SJP wishes to dedicate this paper to the fond memory of her father. SJP acknowledges financial support for this work from a New Zealand Science and Technology Postdoctoral Fellowship and the Ministry of Agriculture and Forestry Sustainable Land Management and Climate Change Plan of Action Programme. We thank Martin Tickner for generously allowing us access to the sites on his property, analytical services provided by Lincoln University, advice from Trevor Webb on the soil properties at the sites, Danny Thornburrow for the analysis of soil physical properties, Guy Forrester and Greg Arnold for statistical assistance, and Kevin Tate for his helpful suggestions to improve the manuscript.

References

Allen DE, Mendham DS, Singh BP, Cowie A, Wang W, Dalai RC, Raison RJ (2009) Nitrous oxide and methane emissions from soil are reduced following afforestation in three contrasting climatic zones. Australian Journal of Soil Research 47, 443-458. doi:10.1071/SR08151

Amundson R (2001) The carbon budget of soil. Annual Review of Earth and Planetary Sciences 29, 535 562. doi:10.1146/annurev.earth.29. 1.535

Anon, (1983) 'Summaries of climatological observations to 1980.' (New Zealand Meteorological Service: Wellington, NZ)

Arneth A, Kelliher FM, McSeveny TM, Byers JN (1998) Fluxes of carbon and water in a Pinus radiata forest subject to soil water deficit. Australian Journal of Plant Physiology 25, 557-570. doi:10.1071/PP97153

Ball BC, McTaggart IP, Watson CA (2002) Influence of organic ley-arable management and afforestation in sandy loam to clay loam soils on fluxes of [N.sub.2]O and C[H.sub.4] in Scotland. Agriculture, Ecosystems & Environment 90, 305-317. doi:10.1016/S0167-8809(01)00207-9

Ball T, Smith KA, Moncrieff JB (2007) Effect of stand age on greenhouse gas fluxes from a Sitka spruce [Picea sitchensis (Bong.) Carr.] chronosequence on a peaty gley soil. Global Change Biology 13, 2128-2142. doi:10.1111/j.1365-2486.2007.01427.x

Blakemore L, Searle P, Daly B (1987) 'Methods for chemical analysis of soils.' New Zealand Soil Bureau Scientific Report No. 80. (New Zealand Soil Bureau: Lower Hutt, NZ)

Borken W, Davidson EA, Savage K, Sundquist ET, Steudler P (2006) Effect of summer throughfall exclusion, summer drought, and winter snow cover on methane fluxes in a temperate forest soil. Soil Biology & Biochemistry 38, 1388-1395. doi:10.1016/j.soilbio.2005.10.011

Brown M, Whitehead D, Hunt JE, Clough TJ, Arnold GC, Baisden WT, Sherlock RR (2008) Regulation of soil-surface respiration in a grazed pasture in New Zealand. Agricultural and Forest Meteorology. doi:10.1016/j.agrformet.2008.08.005

Castaldi S, Fierro A (2005) Soil-atmosphere methane exchange in undisturbed and burned Mediterranean shrubland of southern Italy. Ecosystems 8, 182-190. doi:10.1007/s10021-004-0093-z

Castro MS, Mellilo JM, Steudler PA, Chapman JW (1994) Soil moisture as a predictor of methane uptake by temperate forest soils. Canadian Journal of Forest Research--Revue Canadienne de Recherche Forestiere 24, 1805 1810. doi:10.1139/x94-233

Clough TJ, Price SJ, Sherlock RR (1998) Rapid automated analysis of nitrous oxide, methane and oxygen. In 'Proceedings of the Workshop on the Science of Atmospheric Trace Gases'. Technical Report 15. (Ed. TS Clarkson) pp. 45-47. (NIWA: Wellington, New Zealand)

Conrad R (1996) Soil microorganisms as controlle[R.sub.s] of atmospheric trace gases ([H.sub.2], CO, C[H.sub.4], OCS, [N.sub.2]O and NO). Microbiological Reviews 60, 609-640.

Cook FJ, Orchard VA (2008) Relationships between soil respiration and soil moisture. Soil Biology & Biochemistry 40, 1013-1018. doi:10.1016/ j.soilbio.2007.12.012

Curtis PS, Vogel CS, Gough CM, Schmid HP, Su HB, Bovard BD (2005) Respiratory carbon losses and the carbon-use efficiency of a northern hardwood forest, 1999-2003. New Phytologist 167, 437-456. doi:10.1111/j.1469-8137.2005.01438.x

Dalal RC, Allen DE (2008) Greenhouse gas fluxes from natural ecosystems. Turner Review No. 18. Australian Journal of Botany 56, 369-407. doi:10.1071/BT07128

Davidson EA, Belk E, Boone RD (1998) Soil, water content and temperature as independent or confounded factors controlling soil respiration in a temperate mixed hardwood forest. Global Change Biology 4, 217-227. doi:10.1046/j.1365-2486.1998.00128.x

Davidson EA, Reis de Carvalho CJ, Figueira AM, Ishida FY, Ometto JPHB, Nardoto GB, Saba RT, Hayashi SN, Leal EC, Vieira IMCG, Martinelli LA (2007) Recuperation of nitrogen cycling in Amazonian forests following agricultural abandonment. Nature 447, 995-998. doi:10.1038/nature05900

Del Grosso SJ, Parton WJ, Mosier AR, Ojima DS, Potter CS, Borken W, Brumme R, Butterbach-Bahl K, Crill PM, Dobbie K, Smith KA (2000) General C[H.sub.4] oxidation model and comparisons of C[H.sub.4] oxidation in natural and managed systems. Global Biogeochemical Cycles 14, 999-1019.

Dutaur L, Verchot LV (2007) A global inventory of the soil C[H.sub.4] sink. Global Biogeochemical Cycles 21, GB4013. doi:10.1029/2006GB002734

Erickson H, Keller M, Davidson EA (2001) Nitrogen oxide fluxes and nitrogen cycling during post agricultural succession and forest fertilisation in the humid tropics. Ecosystems 4, 67-84. doi:10.1007/ s 100210000060

Fang Y, Gundersen P, Zhang W, Zhou G, Christiansen JR, Mo J, Dong S, Zhang T (2009) Soil atmosphere exchange of [N.sub.2]O, C[O.sub.2] and C[H.sub.4] along a slope of an evergreen broad-leaf forest in southern China. Plant and Soil 319, 37-48. doi:10.1007/s11104-008-9847-2

Forster P, Ramaswamy V, Artaxo P, Berntsen T, Betts R, Fahey DW, Haywood J, Lean J, Lowe DC, Myhre G, Nganga J, Prinn R, Raga G, Schulz M, Van Dorland R (2007) Changes in atmospheric constituents and in radiative forcing. In 'Climate Change 2007: The Physical Science Basis'. Contribution of Working Group 1 to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. (Eds S Solomon, D Qin, M Manning, Z Chen, M Marquis, KB Averyt, M Tignor, HL Miller) pp. 131-234. (Cambridge University Press: Cambridge, New York)

Harding JS (2003) Historic deforestation and the fate of endemic invertebrate species in streams. New Zealand Journal of Marine and Freshwater Research 37, 333-345. doi:10.1080/00288330.2003.9517171

Hewitt AE (1998) 'New Zealand Soil Classification'. 2nd edn, Landcare Research Science Series 1. (Manaaki-Whenua Press: Lincoln, New Zealand)

Hunt JE, Kelliher FM, McSeveny TM, Byers JN (2002) Evaporation and carbon dioxide exchange between the atmosphere and a tussock grassland during a summer drought. Agricultural and Forest Meteorology 111, 65-82. doi:10.1016/S0168-1923(02)00006-0

Judd M J, Kelliher FM, Ulyatt MJ, Lassey KR, Tate KR, Shelton ID, Harvey MJ, Walker CF (1999) Net methane emissions from grazing sheep. Global Change Biology 5, 647-657. doi:10.1046/j.1365-2486.1999. 00264.x

Kammann C, Hepp S, Lenhart K, Muller C (2009) Stimulation of methane consumption by endogenous C[H.sub.4] production in aerobic grassland soil. Soil Biology & Biochemistry 41, 622-629. doi:10.1016/j.soilbio. 2008.12.025

Kelliher FM, Lloyd J, Arneth A, Luhker B, Byers JN, McSeveny TM, Milukova I, Grigoriev S, Panfyorov M, Sogatchev A, Varlargin A, Ziegler W, Bauer G, Wong S-C, Schulze ED (1999) Carbon dioxide efflux density from the floor of a central Siberian pine forest. Agricultural and Forest Meteorology 94, 217-232. doi:10.1016/S0168-1923(99)00014-3

Klute A (1986) Water retention: laboratory methods. In 'Methods of soil analysis, Part 1'. (Ed. A Klute) pp. 365-662. (ASA-SSSA: Madison, WI)

Law BE, Kelliher FM, Baldocchi DD, Anthoni PM, Irvine J, Moore D, Van Tuyl S (2001) Spatial and temporal variation in respiration in a young ponderosa pine forest during the summer drought. Agricultural and Forest Meteorology 110, 27-43. doi:10.1016/S0168-1923(01)00279-9

Li Y, Xu M, Sun OJ, Cui W (2004) Effects of root and litter exclusion on soil C[O.sub.2] efflux and microbial biomass in wet tropical forests. Soil Biology & Biochemistry, 36, 2111-2114. doi:10.1016/j.soilbio.2004.06.003

Li Z, Kelliher FM (2007) Methane oxidation in freely and poorly drained grassland soils and effects of cattle urine application. Journal of Environmental Quality 36, 1241-1248. doi:10.2134/jeq2006.0237

Livesley SJ, Kiese R, Miehle P, Weston C J, Butterbach-bahl K, Arnt SK (2009) Soil-atmosphere exchange of greenhouse gases in a Eucalyptus marginatu woodland, a clover-grass pasture, and Pinus radiata and Eucalyptus globulus plantations. Global Change Biology 15, 425-440. doi:10.1111/j.1365-2486.2008.01759.x

Lloyd J, Taylor JA (1994) On the temperature dependence of soil respiration. Functional Ecology 8, 315-323. doi:10.2307/2389824

Lowe D (2006) A green source of surprise. Nature 439, 148-149. doi:10.1038/439148a

Luo J, Lindsey SB, Ledgard SF (2008) Nitrous oxide emissions from animal urine application on a New Zealand pasture. Biology and Fertility of Soils 44, 463-470. doi:10.1007/s00374-007-0228-4

Maljanen M, Hytonen J, Martikainen PJ (2001) Fluxes of [N.sub.2]O, C[H.sub.4] and C[O.sub.2] on afforested boreal agricultural soils. Plant and Soil 231, 113-121. doi:10.1023/A:1010372914805

Millard P, Midwood AJ, Hunt JE, Barbour MM, Whitehead D (2010) Quantifying the contribution of soil organic matter to forest soil respiration, using natural abundance [[delta].sup.13]C. Soil Biology & Biochemistry 42, 935-943. doi:10.1016/j.soilbio.2010.02.010

Morishita T, Sakata T, Takahashi M, Ishisuka S, Mizoguchi T, Inagaki Y, Terazawa K, Sawata S, Igarashi M, Yasuda H, Koyama Y, Suzuki Y, Toyota N, Muro M, Kinjo M, Yamamoto H, Ashiya D, Kanazawa Y, Hashimoto T, Umata H (2007) Methane uptake and nitrous oxide emission in Japanese forest soils and their relationship to soil and vegetation types. Soil Science and Plant Nutrition 53, 678-691. doi:10.1111/j.1747-0765.2007.00181.x

Mummey DL, Smith JL, Bolton H Jr (1997) Small-scale spatial and temporal variability of [N.sub.2]O flux from a shrub-steppe ecosystem. Soil Biology & Bioehemistry 29, 1699-1706. doi:10.1016/S0038-0717(97)00077-1

Nieveen JP, Campbell DI, Schipper LA, Blair IJ (2005) Carbon exchange of grazed pasture on a drained peat soil. Global Change Biology 11, 607-618. doi:10.1111/j.1365-2486.2005.00929.x

Parfitt RL, Schipper LA, Baisden WT, Elliott AH (2006) Nitrogen inputs and outputs for New Zealand in 2001 at national and regional scales. Biogeochemistry 80, 71-88. doi:10.1007/s10533-006-0002-y

Pinheiro JC, Bates DM (2000) 'Mixed-effects models in S and S-Plus.' (Springer-Verlag: New York)

Price S J, Sherlock RR, Kelliher FM, McSeveny TM, Tate KR, Condron LM (2004) Pristine New Zealand forest soil is a strong methane sink. Global Change Biology 10, 16-26. doi:10.1046/j.1529-8817.2003.00710x

Prieme A, Christensen S, Dobbie KE, Smith KA (1997) Slow increase in rate of methane oxidation in soils with time following land use change from arable agriculture to woodland. Soil Biology & Biochemistry 29, 1165-1172. doi:10.1016/S0038-0717(97)00038-2

R Development Core Team (2008) 'R: A language and environment for statistical computing.' (R Foundation for Statistical Computing: Vienna, Austria) Available at: www.R-project.org (accessed 9 April 2008).

Reay SD, Norton DA 0999) Assessing the success of restoration plantings in a temperate New Zealand forest. Restoration Ecology 7, 298-308. doi:10.1046/j.1526-100X.1999.72023.x

Risch AC, Frank DA (2006) Carbon dioxide fluxes in a spatially and temporally heterogenous temperate grassland. Oecologia 147, 291-302. doi:10.1007/s00442-005-0261-7

Roslev P, Iverson N, Henriksen K (1997) Oxidation and assimilation of atmospheric methane by soil methane oxidizers. Applied and Environmental Microbiology 63, 874-880.

Ross DJ, Tate KR, Feltham CW (1996) Microbial biomass and C and N mineralisation in litter and mineral soil of adjacent montane ecosystems in a southern beech (Nothofagus) forest and a tussock grassland. Soil Biology & Biochemistry 28, 1613 1620. doi:10.1016/S0038-0717(96)00255-6

Ross DJ, Tate KR, Scott NA, Wilde RH, Rodda NJ (2002) Afforestation of pastures with Pinus radiata influences soil carbon and nitrogen pools and mineralisation and microbial properties. Australian Journal of Soil Research 40, 1303-1318. doi:10.1071/SR02020

Rowe LK, Marden M, Rowan D (1999) Interception and throughfall in a regenerating stand of kanuka (Kunzea ericoides var. ericoides), East Coast Region, North Island, New Zealand, and implications for soil conservation. Journal of Hydrology. New Zealand 38, 29-48.

Saggar S, Tate KR, Giltrap DL, Singh J (2008) Soil-atmosphere exchange of nitrous oxide and methane in New Zealand terrestrial ecosystems and their mitigation options: a review. Plant and Soil 309, 25-42. doi:10.1007/s11104-007-9421-3

Sato A, Seto M (1999) Relationship between carbon dioxide evolution, microbial biomass carbon and amount of dissolved organic carbon as affected by temperature and water content of a forest and amble soil. Communications in Soil Science and Plant Analysis 30, 2593-2605. doi:10.1080/00103629909370399

Smith KA, Dobbie KE, Ball BC, Bakken LR, Sitaula BK, Hansen S, Brumme R, Borken W, Christensen S, Prieme A, Fowler D, MacDonald JA, Skiba U, Klemedtsson L, Kasmir-Klemedtsson A, Degorska A, Orlanski P (2000) Oxidation of atmospheric methane in Northern European soils, comparison with other ecosystems, and uncertainties in the global terrestrial sink. Global Change Biology 6, 791-803. doi:10.1046/j.1365-2486.2000.00356.x

Steudler PA, Melillo JM, Bowden RD, Castro MS, Lugo AE (1991) The effects of natural and human disturbances on soil nitrogen dynamics and trace gas fluxes in a Puerto Rican wet forest. Biotropica 23, 356-363. doi:10.2307/2388252

Steudler PA, Mellilo JM, Feigel BJ, Neill C, Piccolo MC, Cerri CC (1996) Consequence of forest to pasture conversion on C[H.sub.4] fluxes in the Brazilian Amazon Basin. Journal of Geophysical Research Atmospheres 101, 18547-18554. doi:10.1029/96JD01551

Suwanwaree P, Robertson GP (2005) Methane oxidation in forest, successional and no-till agricultural ecosystems: effects of nitrogen and soil disturbance. Soil Science Society of America Journal 69, 1722-1729. doi:10.2136/sssaj2004.0223

Tate KR, Ross DJ, Saggar S, Hedley CB, Dando J, Singh BK, Lambie SM (2007) Methane uptake in soils from Pinus radiata plantations, a reverting shrubland and adjacent pastures: Effects of land use change, and soil texture, water and mineral nitrogen. Soil Biology & Biochemistry 39, 1437 1449. doi:10.1016/j.soilbio.2007.01.005

Tate KR, Ross DJ, Scott NA, Rodda NJ, Townsend JA, Arnold GC (2006) Post-harvest patterns of carbon dioxide production, methane uptake and nitrous oxide production in a Pinus radiata D. Don plantation. Forest Ecology and Management 228, 40-50. doi:10.1016/j.foreco.2006.02.023

Tate KR, Wilde RH, Giltrap DJ, Baisden WT, Saggar S, Trustrum NA, Scott NA, Barton JP (2005) Soil organic carbon stocks and flows in New Zealand: System development, measurement and modelling. Canadian Journal of Soil Science 85, 481-489.

Trangmar BB (1986) Soil-landscape relationships Barrys Bay-Wainui District, Banks Peninsula. New Zealand Soil Bureau District Office Report CH 20. Department of Scientific and Industrial Research, New Zealand.

Trotter CM, Tate KR, Scott N, Townsend J, Wilde H, Lambie S, Marden M, Pinkney T (2005) Afforestation/reforestation of New Zealand marginal pasture lands by indigenous shrublands: The potential for Kyoto forest sinks. Annals of Forest Science 62, 865-871. doi:10.1051/forest:2005077

Ussiri DAN, Lal R, Jarecki MK (2009) Nitrous oxide and methane emissions from long-term tillage under a continuous corn cropping system in Ohio. Soil & Tillage Research 104, 247 255. doi:10.1016/j.still.2009.03.001

van den Pol-van Dasselaar A, van Beusichem ML, Oenema O (1999) Effects of nitrogen input and grazing on methane fluxes of extensively and intensively managed grasslands in the Netherlands. Biology and Fertility of Soils 29, 24-30. doi:10.1007/s003740050520

Verchot LV, Davidson EA, Cattanio JH, Ackerman IL (2000) Land-use change and biogeochemical controls of methane fluxes in soils of Eastern Amazonia. Ecosystems 3, 41 56. doi:10.1007/s100210000009

Wallenstein MD, Peterjohn WT, Schlesinger WH (2006) N fertilization effects]on denitrification and N cycling in an aggrading forest. Ecological Applications 16, 2168-2176. doi:10.1890/1051-0761(2006)01612168:NFEODA]2.0.CO;2

Wilson HD (1987) The vegetation of Banks Peninsula. In 'The Banks Peninsula Landscape Report'. (Ed. CJ Dorsey) (Queen Elizabeth II National Trust: New Zealand)

Zerva A, Mencuccini M (201353 Short-term effects of clear felling on soil C[O.sub.2], C[H.sub.4] and [N.sub.2]O fluxes in a Sitka spruce plantation. Soil Biology & Biochemistry 37, 2025-2036. doi:10.1016/j.soilbio.2005.03.004

Manuscript received 31 August 2009, accepted 14 April 2010

Sally Price (A,B,C), David Whitehead (A), Robert Sherlock (B), Tony McSeveny (A), and Graeme Rogers (A)

(A) Landcare Research, PO Box 40, Lincoln 7640, New Zealand.

(B) Faculty of Agriculture and Life Science, PO Box 84, Lincoln University 7647, New Zealand.

(C) Corresponding author. Emails: sally.price@lincoln.ac.nz; sjpricenz@yahoo.com
Table 1. Mean soil physical (2 samples) and chemical ([ + or -]
standard error in parentheses, n=3) properties for the unimproved
pasture (UP), young Kunzea (YK), and old Kunzea (OK) sites at 3
depths

BD, Bulk density; FC, field capacity (suction 10 kPa); WP,
permanent wilting point (suction 1500kPa). Soil mineral nitrogen
values are presented on a unit dry mass basis

                                        Soil water
                                        content
                                        ([m.sup.3]/
                             Total      [m.sup.3])
Depth         Dry BD        porosity
(mm)      (Mg/[m.sup.3])   (% volume)    FC     WP        pH

                                            UP

0-50           0.87           66.4      0.35   0.15   6.8 (0.01)
50-100         1.05           59.9      0.38   0.17   6.7 (0.02)
100-150        1.07           58.9      0.36   0.18   6.5 (0.02)
                                            YK

0-50           0.66           73.5      0.30   0.13   5.5 (0.02)
50-100         0.95           63.7      0.36   0.19   5.3 (0.01)
100-150        0.95           63.4      0.34   0.18   5.2 (0.01)

                                            OK

0-50           0.61           75.8      0.34   0.17   5.4 (0.01)
50-100         0.98           62.2      0.41   0.22   5.6 (0.01)
100-150        0.99           62.3      0.39   0.22   5.8 (0.01)

          Mineral-N ([micro]g/g)

Depth     N[H.sub.4.
(mm)        sup.+]      N[0.sub.3]

                     UP

0-50      0.24 (0.09)   0.00
50-100    0.17 (0.09)   0.00
100-150   0.06 (0.06)   0.04 (0.04)
                     YK

0-50      2.02 (0.44)   0.53 (0.45)
50-100    1.23 (0.25)   0.37 (0.09)
100-150   0.89 (0.13)   0.46 (0.18)

                     OK

0-50      1.53 (0.09)   2.98 (0.35)
50-100    0.44 (0.12)   1.77 (0.33)
100-150   0.20 (0.11)   0.57 (0.15)

Table 2. Parameters obtained by fitting the data to the repeated
measures mixed effects models described by Eqn 1 for C[O.sub.2]
using soil temperature ([T.sub.s]) and root-zone water content
([theta]) as variables at the 3 sites, unimproved pasture (UP),
young Kunzea (YK), and old Kunzea (OK)

The units for the variables and parameters are [T.sub.s] (K), [theta]
and [[theta].sub.c]. ([m.sup.3]/[m.sup.3]), [R.sub.10]
(gC[O.sub.2]/[m.sup.2] .h), and [E.sub.o] (kJ mol/K). None of the
interactive terms were significant

                                               [[theta].
Site    Variables     [R.sub.10]   [E.sub.o]    sub.c]     P value

UP     [T.sub.s], 0      0.26        348.0       0.29      <0.0001
YK     [T.sub.s], 0      0.29        259.0       0.29      <0.0001
OK     [T.sub.s], 0      0.51        289.9       0.27      <0.0001

Table 3. Parameters obtained by fitting the data to the
repeated-measures mixed effects models described by Eqn 2 for
C[H.sub.4]and [N.sub.2]O using soil temperature ([T.sub.s])
and root-zone water content ([theta]) as variables at the 3 sites,
unimproved pasture (UP), young Kunzea (YK) and old Kunzea (OK)

The units for the variables and parameters are [T.sub.s],
([degrees]C), [theta] ([m.sup.3]/[m.sup.3]), and a
([micro]g/m.sup.2] .h) with [b.sub.1], [b.sub.2], and [b.sub.3]
as dimensionless constants; n.s., not significant; n.d., not
determined

Site        Variables        [b.sub.1]    [b.sub.2]   [b.sub.3]

UP     [T.sub.s], [theta]       2.04       -33.12       n.d.
             [theta]            n.d.       137.23       n.d.
YK     [T.sub.s], [theta]      -2.48       -54.86       n.d.
             [theta]            n.d.        88.02       n.d.
OK     [T.sub.s], [theta]      -3.22        96.35       n.d.
             [theta]            n.d.       156.19       n.d.

                            Nitrous oxide

OK     [T.sub.s] [theta],      -1.00        27.20       5.53
       [T.sub.s] x [theta]

Site      a             P value

              Methane

UP       20.88     <0.001, 0.56 (n.s.)
        -48.72            <0.001
YK       35.06   0.44 (n.s.), 0.80 (n.s.)
        -29.16         0.46 (n.s.)
OK       48.35   0.10 (n.s.), 0.10 (n.s.)
         92.41            <0.001

             Nitrous oxide

OK        6.95      0.00, 0.02, 0.01

Table 4. Estimated annual modelled flux densities for C[O.sub.2],
C[H.sub.4], and [N.sub.2]O at the unimproved pasture (UP), young
Kunzea (YK), and old Kunzea (OK) sites

For comparison, the values for C[H.sub.4] at the YK site are
estimated directly from the measurements made, since the model
could not be used at this site. Mean [+ or -] standard error
values for the C[H.sub.4] fluxes estimated directly from the UP
and OK sites are included in parentheses as well as the
appropriate global warming potentials. Values for C[H.sub.4] and
[N.sub.2]O are also shown as C[O.sub.2]-equivalent amounts using
global warming potentials for C[H.sub.4] and [N.sub.2]O over a
100-year period of 25 and 310, respectively (Forster et al. 2007)

       Carbon dioxide             Methane                  (kg
       (x [10.sup.3]          (kg C[H.sub.4]/          C[O.sub.2]-e/
Site    kg/ha.year)              ha.year)                ha.year)

UP          21.9        1.52 (1.24 [+ or -] 10.29)     38.0 (31.0)
YK          23.3            -0.65 [+ or -] 0.56            -16.3
OK          39.0        5.09 ( 4.73 [+ or -] 0.85)    127.3 (-118.25)

           Nitrous           oxide
       (kg [N.sub.2]O/   C[O.sub.2]-e/
Site      ha.year)         ha.year)

UP          <0.04            <12.4
YK          <0.14            <43.4
OK           0.25             77.5
COPYRIGHT 2010 CSIRO Publishing
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2010 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Price, Sally; Whitehead, David; Sherlock, Robert; McSeveny, Tony; Rogers, Graeme
Publication:Australian Journal of Soil Research
Article Type:Report
Geographic Code:8NEWZ
Date:Aug 1, 2010
Words:10158
Previous Article:Mechanisms for high Cd activity in a red soil from southern China undergoing gradual reduction.
Next Article:Soil microbial respiration responses to changing temperature and substrate availability in fertile grassland.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters