- Email: [email protected]
Spatial variation in respiration from coarse woody debris in a temperate secondary broad-leaved forest in Japan
Available online at www.sciencedirect.com
Forest Ecology and Management 255 (2008) 149–155 www.elsevier.com/locate/foreco
Spatial variation in respiration from coarse woody debris in a temperate secondary broad-leaved forest in Japan Mayuko Jomura a,c,*, Yuji Kominami b, Masako Dannoura a, Yoichi Kanazawa a b
a Graduate School of Science and Technology, Kobe University, 1-1 Rokkodai, Nada, Kobe, Hyogo 657-8501, Japan Kansai Research Center, Forestry and Forest Products Research Institute, 68 Nagaikyutaro, Momoyama, Fushimi, Kyoto 611-0855, Japan c National Institute for Agro-Environmental Science, 3-1-3 Kannondai, Tsukuba, Ibaraki 305-8604, Japan
Received 26 February 2007; received in revised form 28 July 2007; accepted 3 September 2007
Abstract We measured the rates of respiration from snags and logs (‘‘coarse woody debris’’, CWD) of Japanese red pine (Pinus densiflora Sieb. et Zucc.) to examine the rate of decomposition and CO2 efflux from these materials in a temperate secondary broad-leaved forest in Japan. At this site, a high quantity of CWD of P. densiflora had accumulated as a result of pine wilt disease during the 1970s. Respiration rates were measured using a dynamic closed chamber method combined with an infrared gas analyzer. We measured the respiration rate of 7 samples of snags and 10 samples of logs from August 2003 to January 2004. The responses of the respiration rates of snags (Rsnag) and logs (Rlog) to changing temperature were both exponential and the responses to water content were quadratic, and the same function could be used to estimate annual values of both Rsnag and Rlog. Intensive measurements of water contents of snags and logs showed a marked difference in water content. The mean water content of snags was 20% of log water content. This difference was likely responsible for the observed difference in annual Rsnag and Rlog. The decay rate constants estimated from the respiration rates measurement of snags and logs were 0.019 and 0.081 year1, respectively. Despite being lower than Rlog, Rsnag was a significant compartment of the CWD carbon budget at this site. # 2007 Elsevier B.V. All rights reserved. Keywords: Decomposition rate; Environmental factors; Infrared gas analysis; Wood decay
1. Introduction Coarse woody debris (CWD) plays an important role in forest ecosystems. CWD contains a large stock of carbon and nutrients, provides habitat for various microbes and invertebrates, maintains biodiversity, and affects both carbon and nutrient cycling in a forest (Harmon et al., 1986; Hammond et al., 2001; Krankina et al., 2003; Hicks et al., 2003). In recent studies, the amount and structure of CWD, its input and decomposition rates, and its contribution to the forest’s heterotrophic respiration and carbon budgets have been analyzed (Daniels et al., 1997; Siitonen et al., 2000; Chambers et al., 2001; Busing, 2005; Jomura et al., 2007). In addition, studies of CWD dynamics in several forests with different ages (as a result of forest fires) and model descriptions have revealed * Corresponding author at: National Institute for Agro-Environmental Science, 3-1-3 Kannondai, Tsukuba, Ibaraki 305-8604, Japan. Tel.: +81 29 838 8236; fax: +81 29 838 8236. E-mail address: [email protected] (M. Jomura). 0378-1127/$ – see front matter # 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.foreco.2007.09.002
the important contribution of CWD to long-term forest carbon cycles (Janisch and Harmon, 2002; Bond-Lamberty et al., 2003; Gower, 2003). For determination of a forest’s carbon budget, rates of both production of vegetation and decomposition of organic matter must be quantified. CWD is an important component of the decomposing organic matter, so accounting for CWD respiration as a component of heterotrophic respiration is essential to determining a forest’s carbon budget (Chambers et al., 2001; Bond-Lamberty et al., 2003; Howard et al., 2004). Spatial heterogeneity is a major characteristic of CWD in forests (Harmon et al., 1986). A complex spatial distribution arises from the different positions of components of CWD, and specifically those of standing wood (‘‘snags’’, including fallen branches that remain suspended above the ground and leaning dead trees) and fallen wood on the ground (‘‘logs’’). These different positions change the environmental factors that affect CWD and, thus, affect the decomposition mechanisms (Næsset, 1999). Snag decomposition rates are usually slower than those of logs (Harmon et al., 1986; Yatskov et al., 2003). Limited
150
M. Jomura et al. / Forest Ecology and Management 255 (2008) 149–155
contact with the ground and increased exposure to air circulation decrease the water content of snags (Fahey, 1983; Johnson and Greene, 1991), resulting in slower decomposition rates. Microbial invasion is also limited for snags, although the invasion rate varies considerably among tree species (Harmon et al., 1986). To elucidate the landscape-scale forest carbon cycle, the spatial heterogeneity in CWD decomposition rates must thus be examined. CWD decomposition rates have been measured frequently in previous studies (e.g., Harmon et al., 1986; Laiho and Prescott, 2004; Mackensen et al., 2003). Most studies have used the rate of weight loss by CWD as an estimate of decomposition rates. However, weight loss of CWD includes three main processes: losses due to microbial respiration, leaching, and fragmentation. Decomposition rates based on weight loss thus evaluate the combined effects of these different biological, chemical and physical processes. However, the current emphasis on carbon cycles in forest ecosystems is on CO2 fluxes. CWD respiration is a significant component of overall heterotrophic respiration (about 10–16% in a temperate secondary broad-leaved forest in Japan, Jomura et al., 2007) and is thus an important factor in evaluating net ecosystem production (NEP; Chambers et al., 2001) and the contribution of decomposition to CO2 fluxes. Thus, the respiratory loss of carbon from CWD should be evaluated separately from the other two processes to permit a more precise evaluation of NEP in forest ecosystems. The objective of our study was thus to clarify the effect of differences in the spatial positions (standing and downed) of CWD on the respiration rate. We hypothesized that the annual carbon loss from CWD would vary spatially in response to environmental factors. We constructed a function for the relationship between environmental factors and the respiration rates of snags and logs, and evaluated its ability to explain the variation in CWD respiration rates. We also evaluated the validity of the respiration measurement used to estimate decomposition rates of CWD. 2. Site and methods 2.1. Site description The study site was located in a temperate secondary broadleaved forest in central Japan (the Yamashiro Experimental Forest, 348470 N, 1358500 E). More than 100 years ago, there were few trees in the area because of severe cutting of trees for firewood. Trees were planted ca. 100 years ago for soil and water conservation. Subsequently, the area was dominated by Pinus densiflora Sieb. et Zucc. About 30 years ago, pine wilt disease spread throughout this area; most of the P. densiflora died, and broad-leaved species took their place. In 1999, the living tree biomass (DBH 3 cm) was estimated to be 48.3 t C ha1, and deciduous broad-leaved tree species such as Quercus serrata Thunb. ex Murray, evergreen broad-leaved tree species such as Ilex pedunculosa Miq., and coniferous tree species such as P. densiflora accounted for 66, 28, and 6% of the total biomass, respectively (Goto et al., 2003). In 2003, the mass of dead wood (DBH 3 cm) was estimated to be
9.1 t C ha1, and 73% of the total was standing and downed dead wood of P. densiflora (Jomura et al., 2007). The annual mean air temperature was 15.5 8C, and the hourly maximum and minimum air temperatures in 2002 were observed in the summer (34.8 8C) and winter (3.9 8C; Goto et al., 2003). Mean annual precipitation was 1449 mm; the rainy season occurred in late June and early July, and some typhoons contributed significant amounts of precipitation to the study area in the summer and fall. The study site was a 1.7 ha catchment (220 m above sea level), with a mean canopy height of 12.0 m, and mean DBH and tree height of 7.4 cm and 5.6 m, respectively, for woody vegetation with DBH 3 cm. The stand density averaged 3209 trees ha1 and the total basal area averaged 20.7 m2 ha1 (Goto et al., 2003). The forest soil was immature and sandy, was derived from weathered granite, and had thin O and A layers. 3. Methods Because the majority of the CWD at our study site was P. densiflora materials, we focused our analysis on these materials. Samples were obtained from standing P. densiflora (snags, n = 7) and downed dead wood of this species (logs, n = 10). All samples were of similar age (judged based on the fact that the bark remained attached), so we did not attempt to estimate the age of the samples. The diameter and length of each sample were measured and volumes were calculated based on an assumption that samples were regular cylinders. Subsamples were also obtained to estimate the oven-dry wood density. The dry weight of each sample was calculated by multiplying the sample volume by the wood density of the subsample. Snags and logs had similar mean wood densities (0.29 0.05 and 0.30 0.11 g cm3, respectively; P = 0.803) and diameters (16.9 5.6 and 15.4 6.2 cm, respectively, P = 0.785). Samples were cut to a mean length of 69 2 cm to fit within the sample chamber (described below). Cut surfaces of the samples were sealed with a silicone sealant to prevent the invasion of microbes and emission of CO2 at these surfaces. All samples were retained at the study site. Snag samples were stood vertically against a steel frame and were isolated from the ground by a vinyl sheet spread under the frame. Log samples lay directly on the ground. Respiration rates of snag and log samples were measured in a dynamic closed chamber connected to an infrared gas analyzer (IRGA). The system was composed of an LI-800 IRGA meter (LI-COR Inc., Lincoln, NE, USA), a chamber made of acrylic resin (W 30 cm D 80 cm H 30 cm inner diameter; 72 000 cm3), polyethylene tubes, a pump (GS-3EA, Enomoto Micro-Pump, Tokyo, Japan), filters, and a flow meter (Fig. 1). The temperature in the chamber (Tc) was measured using a copper-constantan thermocouple. We enclosed individual snag and log samples in the chamber, and measured the CO2 concentration in the chamber for 10 min. The CO2 concentration and temperature in the chamber were recorded each second using a datalogger (NR-1000, Keyence, Osaka, Japan). To avoid the effects of air disturbance caused by opening the chamber cover, the data for the 60-s period after the
M. Jomura et al. / Forest Ecology and Management 255 (2008) 149–155
151
Fig. 1. Diagram of the design of the closed-dynamic chamber used to measure respiration from coarse woody debris.
wood densities of these snags and logs were 0.65 and 0.21 g cm3, respectively. Temperature and water content of the log were measured from January to December 2003 and those of the snag were measured from July to December 2003. We measured the volumetric water content of the snag and log samples by means of TDR and converted them to volumetric water contents based on the void volume. Nonlinear regression of the relationships between air temperature and RCWD and between wood moisture content and RCWD were conducted using the IMSL software (Visual Numerics, Houston, TX, USA). We estimated annual carbon loss of snags and logs using the function for the relationships between temperature, water content, and CWD respiration. To estimate the annual CWD respiration rate, we used hourly temperature and water content data for the monitored snag and log. The decay rate constant (k, year1) was obtained using the single exponential model of Olson (1963):
start of the measurements and for the 60-s period before the end of each measurement period were not used.
CWt0 CECWD k ¼ ln CWt0
RCWD ¼ DCO2
V Vs Ta M CO2 103 602 V air T a þ T cell 103 W
(1)
The CWD respiration rate (RCWD, mg CO2 kg1 h1) was calculated using Eq. (1), where DCO2 is the mean rate of change of the CO2 concentration in the chamber (ppm), V the volume of the system (8.08 L, including the chamber volume and the volume of the tubing), Vs the volume of the CWD sample (L), Vair the standard gas volume (22.41 L), Ta the difference between Celsius and absolute temperature (273.2), Tcell the temperature in the IRGA (8C), MCO2 the molecular weight of CO2 (44.01 g), and W is the dry weight of the CWD sample (g). We measured snag and log respiration rates 11 times between August 2003 and January 2004. Air temperature in the chamber (Tc) was averaged for each measurement period. The water content of the samples was estimated from the difference between the fresh weight at the time of measurement and the dry weight of the sample (estimated by multiplying the sample volume by the subsample density). We calculated a volumetric water content based on the void volume of CWD (upore). Boddy (1983) indicated that gravimetric and volumetric water contents differed among woods with different densities, and that upore was thus more appropriate for evaluating the water content of decaying wood. We calculated upore (m3 m3) using the following equation: upore ¼ V water ðV sample DWsample r1 wood Þ
1
(2)
where Vwater is the water volume (m3), Vsample the sample volume (m3), DWsample the sample dry weight (g), and rwood is the wood substrate density (1.5 g cm3; provided by Asano (1982), for all tree species). We chose one snag and one log sample to measure temporal variations in temperature and water content to a depth of 3 cm using time-domain reflectometry (TDR; Hydra Soil Moisture Probe, Stevens Vitel Inc., Beaverton, OR, USA). These samples were 26 cm in diameter and 75 cm in length. The oven-dry
(3)
where CWt0 is the carbon weight (g C) of the CWD (snag or log) sample at initial time t0 if we assume carbon content to be 0.5, and CECWD is the annual respiratory carbon loss of the CWD (snag or log) sample (g C year1). 4. Results 4.1. Seasonal changes in snag and log respiration rates Means of Rsnag and Rlog of 11 measurements were 15.4 (S.D.: 14.1) and 34.4 (S.D.: 23.4) mg CO2 kg1 h1, respectively. The mean Rlog was consistently higher than the mean Rsnag (paired ttest, P < 0.05). Daily mean Rsnag and Rlog ranged from 0.6 to 31.1 mg CO2 kg1 h1 and from 2.6 to 63.4 mg CO2 kg1 h1, respectively (Fig. 2). Rlog varied substantially with season, with the lowest values in winter and the highest values in summer. However, the highest value of Rsnag was observed in the fall. Temperature in the chamber (Tc) varied seasonally, with the highest values in summer and the lowest values in winter, but did not differ markedly between the snag and log samples. The water content of snags (usnag) and logs (ulog) showed no clear seasonal changes, but ulog was consistently higher than usnag throughout the measurement period (paired t-test, P < 0.05). 4.2. Responses of Rsnag and Rlog to temperature and water content Rsnag and Rlog were positively correlated with temperature (Fig. 3a). The relationship between temperature and respiration rate was best described by an exponential function. As shown by the relationship between the respiration rate standardized at a temperature of 15 8C and the water content (Fig. 3b), variations in respiration rate at a given temperature appear to have been caused mainly by variations in water content. To analyze this relationship, we adopted the quadratic regression
152
M. Jomura et al. / Forest Ecology and Management 255 (2008) 149–155
Fig. 4. Comparison of the observed and calculated respiration rates of snags and logs based on the calculated regression equation [RCWD ¼ 17:614 e0:068T c ðupore þ 0:010Þð1:299 upore Þ2:023 , r2 = 0.61] as a function of temperature and water content.
The residuals of Rsnag and Rlog from the regression of measured versus predicted respiration rates did not differ significantly (Fig. 4). Fig. 2. (a) Respiration rate (RCWD), (b) chamber temperature (Tc), and (c) water content (upore, based on the void volume) of the snag and log samples from August 2003 to January 2004.
curve between water content and respiration proposed by Mielnick and Dugas (2000) because a similar regression curve was attained in our previous study using an automated chamber system (Jomura et al., 2005a). The following regression explained 61% of the variation in Rsnag and Rlog (Fig. 4): RCWD ¼ 17:614 e0:068T c ðupore þ 0:010Þð1:299 upore Þ2:023
(4)
4.3. Time trends in temperature and water content of the snags and logs Temperatures of snags, logs, air, and soil (Tsnag, Tlog, Tair, and Tsoil, the soil temperature at a 20-cm depth) during the measurement period ranged from 1.2 to 35.5 8C, 0.7 to 31.2 8C, 0.0 to 33.8 8C, and 3.0 to 26.5 8C, respectively (Fig. 5a). The temperatures of snags and logs showed diurnal changes that corresponded to changes in air temperature. The relationship between Tsnag and Tlog was obtained by linear regression
Fig. 3. Relationships (a) between chamber temperature (Tc) and respiration rate (RCWD) and (b) between water content (based on void volume, upore) and respiration rate, standardized at a temperature of 15 8C, for the snag and log samples.
M. Jomura et al. / Forest Ecology and Management 255 (2008) 149–155
Fig. 5. Half-hourly (a) temperature and (b) water contents of the snag and log samples. Symbols and range bars show the mean observed values and their standard deviations.
(Tsnag = 1.7677 + 1.0145 Tlog, r2 = 0.97). usnag and ulog varied with precipitation and drying and showed similar sawtooth patterns (Fig. 5b). However, usnag was about 20% of ulog based on a linear regression between the two parameters (usnag = 0.0353 + 0.1907ulog, r2 = 0.50). 4.4. Estimated annual Rsnag and Rlog We estimated Rsnag and Rlog using the functions based on temperature and water content described above (Eq. (4)) and the measured temperatures and water contents of the snags and logs. Rsnag and Rlog values ranged from 0.0 to 15.7 and 3.1 to 38.3 mg CO2 kg1 h1 (with corresponding means of 4.0 and 16.3 mg CO2 kg1 h1), respectively. Both respirations showed clear seasonal changes caused by temperature changes. Annual carbon losses from snags and logs were estimated to be 9.6 and 39.0 g C kg1 year1, respectively; thus, carbon losses from snags were roughly 25% of those from logs. As a result, the decay rate constants for snags and logs based on the single exponential decay model were 0.02 and 0.08 year1, respectively. 5. Discussion The relationship between RCWD and temperature was exponential (Fig. 3a), which was consistent with the results of many other respiration studies (e.g., Kirschbaum, 1995). The variation in RCWD at the same temperature was likely to be
153
caused by differences in water content. RCWD decreased at low and high water contents and reached a maximum at intermediate water contents (Fig. 3b). Two types of relationship between water content and RCWD have been reported in previous studies. Some studies reported that RCWD reached a maximum at an intermediate water content, as in our study (Yoneda, 1980; Jomura et al., 2005a), whereas others reported that RCWD did not decrease, even at high water contents (Chambers et al., 2001; Wang et al., 2002; Bond-Lamberty et al., 2003). This difference may result from the use of different units for water content. In the present study, we defined water content based on the void volume of the CWD (upore) whereas the studies that produced different results used water content based on the dry weight of CWD (udw). udw has a different maximum value, which depends on wood density. Even if the value of udw were the same, the actual quantity of water would depend on pore space. Boddy (1983) indicated that udw should not be used to compare udw between woods with different densities, but did not evaluate the effect of water on microbes. She proposed that measurement of the water in available airfilled pores in wood was effective. Yoneda (1980) categorized the relationships between RCWD and udw based on wood density and indicated that the maximum value of RCWD occurred at intermediate values of udw for each different density of CWD. Thus, if RCWD is modeled based on water content changes over a wide range of wood densities, we should use the water content based on the void volume of the CWD. The responses of Rsnag to changes in temperature and water content were similar to those of Rlog (Fig. 3), and the regression function explained 61% of the variation in Rsnag and Rlog (Fig. 4). This result suggests that microbial development in snags was not largely different from that in logs and that we could use the same function to estimate both respirations at our study site. The isolation of snags from the ground inhibits colonization by microbes, and this may be a primary factor responsible for the slower decomposition of snags (i.e., for the lower respiration rates). Our results showed that even snags could be decomposed if sufficient time had passed (about 30 years after the spread of pine wilt disease) for them to be invaded by microbes. The responses of Rsnag to two environmental factors (temperature and moisture content) are thus also useful for estimating CO2 efflux from CWD, including snags, at a landscape level. Respiration rates estimated using the function derived in this study (Eq. (4)) and intensive collection of environmental data for snags and logs allowed us to reproduce the observed data with our function (Fig. 6). This indicated that the difference in water content between the snags and logs was one of the main factors that controlled the differences between Rsnag and Rlog. However, as seen in the relationships between temperature, water content, and respiration, large variation in respiration data is observed even under the same environmental conditions (Fig. 3). This would reflect differences in microbial activity in the CWD. Microbial activity depends on the substrate chemical quality. Levels of nitrogen and recalcitrant materials such as lignin control the rate of CWD decomposition (e.g., Harmon et al., 1986; Berg and McClaugherty, 2003). We have previously tried to quantify
154
M. Jomura et al. / Forest Ecology and Management 255 (2008) 149–155
Fig. 6. Seasonal changes in estimated and observed Rlog and Rsnag and cumulative carbon loss from snags and logs. Estimates are based on the equation for respiration as a function of temperature and water content [RCWD ¼ 17:614 e0:068T CWD ðupore þ 0:010Þð1:299 upore Þ2:023 ] for the snag and log samples.
microbial biomass in CWD using a fumigation–extraction method to evaluate the relationship between microbial biomass and CWD respiration rate (Jomura et al., 2004). However, continuous measurements were impossible because destructive sampling is required in this method. Finding a non-destructive means of continuously observing and modeling microbial activity would permit more accurate estimates of the pattern of CWD respiration in the future. Annual rates of carbon loss from snags and logs were estimated to be 9.6 and 39.0 g C kg1 year1 (Fig. 6), resulting in decay rate constants of 0.019 and 0.081 year1, respectively. Jomura et al. (2005b) reported decay rate constants of 0.028 and 0.149 year1, respectively, for blocks of P. densiflora wood located above and at the surface of the forest floor at the same site. In this older experiment, weight loss due to fragmentation was excluded, so the decay rate constants expressed decomposition by mineralization (i.e., respiration loss) only. The decay rates of the block specimens were faster than those of the CWD respiration samples in the present study. This difference probably resulted from difference in the size of the sample of CWD; the block specimens were 3 cm on each side, whereas the mean size of the CWD samples in the present study was about 15 cm. According to Jomura et al. (2007), the respiration rate of 3-cm CWD would be about two times that of 15-cm CWD. Therefore, it is fair to say that the decay rate calculated using the annual respiration rate estimated from daily and seasonal changes in environmental factors almost fits the values calculated from the weight loss rate. The estimation of decay rates of organic material based on respiration measurements and function that incorporate the main factors that control decay is thus an appropriate approach for considering shortterm effects of measured environmental factors. 6. Conclusions We determined spatial variation in respiration from CWD (standing and downed) in a temperate secondary broad-leaved forest in Japan. The function that described this relationship by incorporating only two environmental factors (temperature and
moisture content) could explain 61% of the variation in both snag and log respiration rates, but a large amount of variation was still observed under the same environmental conditions. Thus, to determine the absolute value of CWD respiration rate, it will be necessary to incorporate both environmental factors and biological factors such as microbial activity and biomass. The estimation of decay rates of CWD, including snags, based on respiration measurements is an appropriate and effective approach for taking into account the effects of environmental factors and evaluating carbon budgets based on CO2 in forest ecosystems. References Asano, I., 1982. Wood Dictionary. Asakura Press, Tokyo (in Japanese). Berg, B., McClaugherty, C., 2003. Plant Litter. Springer-Verlag, Berlin. Boddy, L., 1983. Microclimate and moisture dynamics of wood decomposing in terrestrial ecosystems. Soil Biol. Biochem. 15, 149–157. Bond-Lamberty, B., Wang, C., Gower, S.T., 2003. Annual carbon flux from woody debris for a boreal black spruce fire chronosequence. J. Geophys. Res. 108 (D3), 8220. Busing, R.T., 2005. Tree mortality, canopy turnover, and woody detritus in old cove forests of the southern Appalachians. Ecology 86, 73–84. Chambers, J.Q., Schimel, J.P., Nobre, A.D., 2001. Respiration from coarse wood litter in central Amazon forests. Biogeochemistry 52, 115–131. Daniels, L.D., Dobry, K., Klinka, K., Feller, M.C., 1997. Determining year of death of logs and snags of Thuja plicate in southwestern coastal British Columbia. Can. J. For. Res. 27, 1132–1141. Fahey, T.J., 1983. Nutrient dynamics of aboveground detritus in lodgepole pine (Pinus contorta ssp. latifolia) ecosystems, southeastern Wyoming. Ecol. Monogr. 53, 51–72. Goto, Y., Kominami, Y., Miyama, T., Tamai, K., Kanazawa, Y., 2003. Aboveground biomass and net primary production of a broad-leaved secondary forest in the southern part of Kyoto prefecture, central Japan. Bull. For. For. Prod. Res. Inst. 387, 115–147. Gower, S.T., 2003. Patterns and mechanisms of the forest carbon cycle. Annu. Rev. Environ. Res. 28, 169–204. Hammond, H.E.J., Langor, D.W., Spence, J.R., 2001. Early colonization of Populus wood by saproxylic beetles (Coleoptera). Can. J. For. Res. 31, 1175–1183. Harmon, M.E., Franklin, J.F., Awanson, F.J., Sollins, P., Gregory, S.V., Lattin, J.D., Anderson, N.H., Cline, S.P., Aumen, N.G., Sedell, J.R., Lienkaemper, G.W., Cromack Jr., K., Cummins, K.W., 1986. Ecology of coarse woody debris in temperate ecosystems. Adv. Ecol. Res. 15, 133–302.
M. Jomura et al. / Forest Ecology and Management 255 (2008) 149–155 Hicks, W.T., Harmon, M.E., Myrold, D.D., 2003. Substrate controls on nitrogen fixation and respiration in woody debris from the Pacific Northwest USA. For. Ecol. Manage. 176, 25–35. Howard, E.A., Gower, S.T., Foley, J.A., Kucharik, C.J., 2004. Effects of logging on carbon dynamics of a jack pine forest in Saskatchewan, Canada. Global Change Biol. 10, 1267–1284. Janisch, J.E., Harmon, M.E., 2002. Successional changes in live and dead wood carbon stores: implications for net ecosystem productivity. Tree Physiol. 22, 77–89. Johnson, E.A., Greene, D.F., 1991. A method for studying dead bole dynamics in Pinus contorta var. latifolia–Picea engelmannii forests. J. Veg. Sci. 2, 523–530. Jomura, M., Kimura, H., Furusawa, H., Kominami, Y., Kanazawa, Y., 2004. The measurement of microbial biomass C in coarse woody debris by fumigation– extraction method. Memoirs of Graduate School of Science and Technology Kobe University 22B, 67–70 (in Japanese with English summary). Jomura, M., Kominami, Y., Kanazawa, Y., 2005a. Long-term measurements of the CO2 flux from coarse woody debris using an automated chamber system. J. Jpn. For. Sci. 87, 138–144 (in Japanese with English summary). Jomura, M., Takeda, F., Kominami, Y., Kanazawa, Y., Goto, Y., 2005b. Spatial variation in the decomposition rate of block specimens of Japanese red pine (Pinus densiflora Sieb. et Zucc.). Appl. For. Sci. 14, 83–88. Jomura, M., Kominami, Y., Tamai, K., Miyama, T., Goto, Y., Dannoura, M., Kanazawa, Y., 2007. The carbon budget of coarse woody debris in a temperate broad-leaved secondary forest in Japan. Tellus 59B, 211–222. Kirschbaum, M.U.F., 1995. The temperature dependence of soil organic matter decomposition, and the effect of global warming on soil organic C storage. Soil Biol. Biochem. 27, 753–760.
155
Krankina, O.N., Harmon, M.E., Kukuev, Y.A., Treyfeld, R.F., Kashpor, N.N., Kresnov, V.G., Skudin, V.M., Protasov, N.A., Yatskov, M., Spycher, G., Povarov, E.D., 2003. Coarse woody debris in forest regions of Russia. Can. J. For. Res. 32, 768–778. Laiho, R., Prescott, C.E., 2004. Decay and nutrient dynamics of coarse woody debris in northern coniferous forests: a synthesis. Can. J. For. Res. 34, 763– 777. Mackensen, J., Bauhus, J., Webber, E., 2003. Decomposition rates of coarse woody debris—a review with particular emphasis on Australian tree species. Aust. J. Bot. 51, 27–37. Mielnick, P.C., Dugas, W.A., 2000. Soil CO2 flux in a tallgrass prairie. Soil Biol. Biochem. 32, 221–228. Næsset, E., 1999. Decomposition rate constants of Picea abies logs in southeastern Norway. Can. J. For. Res. 29, 372–381. Olson, J.S., 1963. Energy storage and the balance of procedures and decomposers in ecological systems. Ecology 44, 322–331. Siitonen, J., Martikainen, P., Punttila, P., Rahh, J., 2000. Coarse woody debris and stand characteristics in mature managed and old-growth boreal mesic forests in southern Finland. For. Ecol. Manage. 128, 211–225. Wang, C., Bond-Lamberty, B., Gower, S.T., 2002. Environmental controls on carbon dioxide flux from black spruce coarse woody debris. Oecologia 132, 374–381. Yatskov, M., Harmon, M.E., Krankina, O.N., 2003. A chronosequence of wood decomposition in the boreal forests on Russia. Can. J. For. Res. 33, 1211– 1226. Yoneda, T., 1980. Studies on the rate of decay of wood litter on the forest floor. 3. Effect of moisture content on CO2 evolution from decaying wood. Jpn. J. Ecol. 30, 55–62.
Forest Ecology and Management 255 (2008) 149–155 www.elsevier.com/locate/foreco
Spatial variation in respiration from coarse woody debris in a temperate secondary broad-leaved forest in Japan Mayuko Jomura a,c,*, Yuji Kominami b, Masako Dannoura a, Yoichi Kanazawa a b
a Graduate School of Science and Technology, Kobe University, 1-1 Rokkodai, Nada, Kobe, Hyogo 657-8501, Japan Kansai Research Center, Forestry and Forest Products Research Institute, 68 Nagaikyutaro, Momoyama, Fushimi, Kyoto 611-0855, Japan c National Institute for Agro-Environmental Science, 3-1-3 Kannondai, Tsukuba, Ibaraki 305-8604, Japan
Received 26 February 2007; received in revised form 28 July 2007; accepted 3 September 2007
Abstract We measured the rates of respiration from snags and logs (‘‘coarse woody debris’’, CWD) of Japanese red pine (Pinus densiflora Sieb. et Zucc.) to examine the rate of decomposition and CO2 efflux from these materials in a temperate secondary broad-leaved forest in Japan. At this site, a high quantity of CWD of P. densiflora had accumulated as a result of pine wilt disease during the 1970s. Respiration rates were measured using a dynamic closed chamber method combined with an infrared gas analyzer. We measured the respiration rate of 7 samples of snags and 10 samples of logs from August 2003 to January 2004. The responses of the respiration rates of snags (Rsnag) and logs (Rlog) to changing temperature were both exponential and the responses to water content were quadratic, and the same function could be used to estimate annual values of both Rsnag and Rlog. Intensive measurements of water contents of snags and logs showed a marked difference in water content. The mean water content of snags was 20% of log water content. This difference was likely responsible for the observed difference in annual Rsnag and Rlog. The decay rate constants estimated from the respiration rates measurement of snags and logs were 0.019 and 0.081 year1, respectively. Despite being lower than Rlog, Rsnag was a significant compartment of the CWD carbon budget at this site. # 2007 Elsevier B.V. All rights reserved. Keywords: Decomposition rate; Environmental factors; Infrared gas analysis; Wood decay
1. Introduction Coarse woody debris (CWD) plays an important role in forest ecosystems. CWD contains a large stock of carbon and nutrients, provides habitat for various microbes and invertebrates, maintains biodiversity, and affects both carbon and nutrient cycling in a forest (Harmon et al., 1986; Hammond et al., 2001; Krankina et al., 2003; Hicks et al., 2003). In recent studies, the amount and structure of CWD, its input and decomposition rates, and its contribution to the forest’s heterotrophic respiration and carbon budgets have been analyzed (Daniels et al., 1997; Siitonen et al., 2000; Chambers et al., 2001; Busing, 2005; Jomura et al., 2007). In addition, studies of CWD dynamics in several forests with different ages (as a result of forest fires) and model descriptions have revealed * Corresponding author at: National Institute for Agro-Environmental Science, 3-1-3 Kannondai, Tsukuba, Ibaraki 305-8604, Japan. Tel.: +81 29 838 8236; fax: +81 29 838 8236. E-mail address: [email protected] (M. Jomura). 0378-1127/$ – see front matter # 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.foreco.2007.09.002
the important contribution of CWD to long-term forest carbon cycles (Janisch and Harmon, 2002; Bond-Lamberty et al., 2003; Gower, 2003). For determination of a forest’s carbon budget, rates of both production of vegetation and decomposition of organic matter must be quantified. CWD is an important component of the decomposing organic matter, so accounting for CWD respiration as a component of heterotrophic respiration is essential to determining a forest’s carbon budget (Chambers et al., 2001; Bond-Lamberty et al., 2003; Howard et al., 2004). Spatial heterogeneity is a major characteristic of CWD in forests (Harmon et al., 1986). A complex spatial distribution arises from the different positions of components of CWD, and specifically those of standing wood (‘‘snags’’, including fallen branches that remain suspended above the ground and leaning dead trees) and fallen wood on the ground (‘‘logs’’). These different positions change the environmental factors that affect CWD and, thus, affect the decomposition mechanisms (Næsset, 1999). Snag decomposition rates are usually slower than those of logs (Harmon et al., 1986; Yatskov et al., 2003). Limited
150
M. Jomura et al. / Forest Ecology and Management 255 (2008) 149–155
contact with the ground and increased exposure to air circulation decrease the water content of snags (Fahey, 1983; Johnson and Greene, 1991), resulting in slower decomposition rates. Microbial invasion is also limited for snags, although the invasion rate varies considerably among tree species (Harmon et al., 1986). To elucidate the landscape-scale forest carbon cycle, the spatial heterogeneity in CWD decomposition rates must thus be examined. CWD decomposition rates have been measured frequently in previous studies (e.g., Harmon et al., 1986; Laiho and Prescott, 2004; Mackensen et al., 2003). Most studies have used the rate of weight loss by CWD as an estimate of decomposition rates. However, weight loss of CWD includes three main processes: losses due to microbial respiration, leaching, and fragmentation. Decomposition rates based on weight loss thus evaluate the combined effects of these different biological, chemical and physical processes. However, the current emphasis on carbon cycles in forest ecosystems is on CO2 fluxes. CWD respiration is a significant component of overall heterotrophic respiration (about 10–16% in a temperate secondary broad-leaved forest in Japan, Jomura et al., 2007) and is thus an important factor in evaluating net ecosystem production (NEP; Chambers et al., 2001) and the contribution of decomposition to CO2 fluxes. Thus, the respiratory loss of carbon from CWD should be evaluated separately from the other two processes to permit a more precise evaluation of NEP in forest ecosystems. The objective of our study was thus to clarify the effect of differences in the spatial positions (standing and downed) of CWD on the respiration rate. We hypothesized that the annual carbon loss from CWD would vary spatially in response to environmental factors. We constructed a function for the relationship between environmental factors and the respiration rates of snags and logs, and evaluated its ability to explain the variation in CWD respiration rates. We also evaluated the validity of the respiration measurement used to estimate decomposition rates of CWD. 2. Site and methods 2.1. Site description The study site was located in a temperate secondary broadleaved forest in central Japan (the Yamashiro Experimental Forest, 348470 N, 1358500 E). More than 100 years ago, there were few trees in the area because of severe cutting of trees for firewood. Trees were planted ca. 100 years ago for soil and water conservation. Subsequently, the area was dominated by Pinus densiflora Sieb. et Zucc. About 30 years ago, pine wilt disease spread throughout this area; most of the P. densiflora died, and broad-leaved species took their place. In 1999, the living tree biomass (DBH 3 cm) was estimated to be 48.3 t C ha1, and deciduous broad-leaved tree species such as Quercus serrata Thunb. ex Murray, evergreen broad-leaved tree species such as Ilex pedunculosa Miq., and coniferous tree species such as P. densiflora accounted for 66, 28, and 6% of the total biomass, respectively (Goto et al., 2003). In 2003, the mass of dead wood (DBH 3 cm) was estimated to be
9.1 t C ha1, and 73% of the total was standing and downed dead wood of P. densiflora (Jomura et al., 2007). The annual mean air temperature was 15.5 8C, and the hourly maximum and minimum air temperatures in 2002 were observed in the summer (34.8 8C) and winter (3.9 8C; Goto et al., 2003). Mean annual precipitation was 1449 mm; the rainy season occurred in late June and early July, and some typhoons contributed significant amounts of precipitation to the study area in the summer and fall. The study site was a 1.7 ha catchment (220 m above sea level), with a mean canopy height of 12.0 m, and mean DBH and tree height of 7.4 cm and 5.6 m, respectively, for woody vegetation with DBH 3 cm. The stand density averaged 3209 trees ha1 and the total basal area averaged 20.7 m2 ha1 (Goto et al., 2003). The forest soil was immature and sandy, was derived from weathered granite, and had thin O and A layers. 3. Methods Because the majority of the CWD at our study site was P. densiflora materials, we focused our analysis on these materials. Samples were obtained from standing P. densiflora (snags, n = 7) and downed dead wood of this species (logs, n = 10). All samples were of similar age (judged based on the fact that the bark remained attached), so we did not attempt to estimate the age of the samples. The diameter and length of each sample were measured and volumes were calculated based on an assumption that samples were regular cylinders. Subsamples were also obtained to estimate the oven-dry wood density. The dry weight of each sample was calculated by multiplying the sample volume by the wood density of the subsample. Snags and logs had similar mean wood densities (0.29 0.05 and 0.30 0.11 g cm3, respectively; P = 0.803) and diameters (16.9 5.6 and 15.4 6.2 cm, respectively, P = 0.785). Samples were cut to a mean length of 69 2 cm to fit within the sample chamber (described below). Cut surfaces of the samples were sealed with a silicone sealant to prevent the invasion of microbes and emission of CO2 at these surfaces. All samples were retained at the study site. Snag samples were stood vertically against a steel frame and were isolated from the ground by a vinyl sheet spread under the frame. Log samples lay directly on the ground. Respiration rates of snag and log samples were measured in a dynamic closed chamber connected to an infrared gas analyzer (IRGA). The system was composed of an LI-800 IRGA meter (LI-COR Inc., Lincoln, NE, USA), a chamber made of acrylic resin (W 30 cm D 80 cm H 30 cm inner diameter; 72 000 cm3), polyethylene tubes, a pump (GS-3EA, Enomoto Micro-Pump, Tokyo, Japan), filters, and a flow meter (Fig. 1). The temperature in the chamber (Tc) was measured using a copper-constantan thermocouple. We enclosed individual snag and log samples in the chamber, and measured the CO2 concentration in the chamber for 10 min. The CO2 concentration and temperature in the chamber were recorded each second using a datalogger (NR-1000, Keyence, Osaka, Japan). To avoid the effects of air disturbance caused by opening the chamber cover, the data for the 60-s period after the
M. Jomura et al. / Forest Ecology and Management 255 (2008) 149–155
151
Fig. 1. Diagram of the design of the closed-dynamic chamber used to measure respiration from coarse woody debris.
wood densities of these snags and logs were 0.65 and 0.21 g cm3, respectively. Temperature and water content of the log were measured from January to December 2003 and those of the snag were measured from July to December 2003. We measured the volumetric water content of the snag and log samples by means of TDR and converted them to volumetric water contents based on the void volume. Nonlinear regression of the relationships between air temperature and RCWD and between wood moisture content and RCWD were conducted using the IMSL software (Visual Numerics, Houston, TX, USA). We estimated annual carbon loss of snags and logs using the function for the relationships between temperature, water content, and CWD respiration. To estimate the annual CWD respiration rate, we used hourly temperature and water content data for the monitored snag and log. The decay rate constant (k, year1) was obtained using the single exponential model of Olson (1963):
start of the measurements and for the 60-s period before the end of each measurement period were not used.
CWt0 CECWD k ¼ ln CWt0
RCWD ¼ DCO2
V Vs Ta M CO2 103 602 V air T a þ T cell 103 W
(1)
The CWD respiration rate (RCWD, mg CO2 kg1 h1) was calculated using Eq. (1), where DCO2 is the mean rate of change of the CO2 concentration in the chamber (ppm), V the volume of the system (8.08 L, including the chamber volume and the volume of the tubing), Vs the volume of the CWD sample (L), Vair the standard gas volume (22.41 L), Ta the difference between Celsius and absolute temperature (273.2), Tcell the temperature in the IRGA (8C), MCO2 the molecular weight of CO2 (44.01 g), and W is the dry weight of the CWD sample (g). We measured snag and log respiration rates 11 times between August 2003 and January 2004. Air temperature in the chamber (Tc) was averaged for each measurement period. The water content of the samples was estimated from the difference between the fresh weight at the time of measurement and the dry weight of the sample (estimated by multiplying the sample volume by the subsample density). We calculated a volumetric water content based on the void volume of CWD (upore). Boddy (1983) indicated that gravimetric and volumetric water contents differed among woods with different densities, and that upore was thus more appropriate for evaluating the water content of decaying wood. We calculated upore (m3 m3) using the following equation: upore ¼ V water ðV sample DWsample r1 wood Þ
1
(2)
where Vwater is the water volume (m3), Vsample the sample volume (m3), DWsample the sample dry weight (g), and rwood is the wood substrate density (1.5 g cm3; provided by Asano (1982), for all tree species). We chose one snag and one log sample to measure temporal variations in temperature and water content to a depth of 3 cm using time-domain reflectometry (TDR; Hydra Soil Moisture Probe, Stevens Vitel Inc., Beaverton, OR, USA). These samples were 26 cm in diameter and 75 cm in length. The oven-dry
(3)
where CWt0 is the carbon weight (g C) of the CWD (snag or log) sample at initial time t0 if we assume carbon content to be 0.5, and CECWD is the annual respiratory carbon loss of the CWD (snag or log) sample (g C year1). 4. Results 4.1. Seasonal changes in snag and log respiration rates Means of Rsnag and Rlog of 11 measurements were 15.4 (S.D.: 14.1) and 34.4 (S.D.: 23.4) mg CO2 kg1 h1, respectively. The mean Rlog was consistently higher than the mean Rsnag (paired ttest, P < 0.05). Daily mean Rsnag and Rlog ranged from 0.6 to 31.1 mg CO2 kg1 h1 and from 2.6 to 63.4 mg CO2 kg1 h1, respectively (Fig. 2). Rlog varied substantially with season, with the lowest values in winter and the highest values in summer. However, the highest value of Rsnag was observed in the fall. Temperature in the chamber (Tc) varied seasonally, with the highest values in summer and the lowest values in winter, but did not differ markedly between the snag and log samples. The water content of snags (usnag) and logs (ulog) showed no clear seasonal changes, but ulog was consistently higher than usnag throughout the measurement period (paired t-test, P < 0.05). 4.2. Responses of Rsnag and Rlog to temperature and water content Rsnag and Rlog were positively correlated with temperature (Fig. 3a). The relationship between temperature and respiration rate was best described by an exponential function. As shown by the relationship between the respiration rate standardized at a temperature of 15 8C and the water content (Fig. 3b), variations in respiration rate at a given temperature appear to have been caused mainly by variations in water content. To analyze this relationship, we adopted the quadratic regression
152
M. Jomura et al. / Forest Ecology and Management 255 (2008) 149–155
Fig. 4. Comparison of the observed and calculated respiration rates of snags and logs based on the calculated regression equation [RCWD ¼ 17:614 e0:068T c ðupore þ 0:010Þð1:299 upore Þ2:023 , r2 = 0.61] as a function of temperature and water content.
The residuals of Rsnag and Rlog from the regression of measured versus predicted respiration rates did not differ significantly (Fig. 4). Fig. 2. (a) Respiration rate (RCWD), (b) chamber temperature (Tc), and (c) water content (upore, based on the void volume) of the snag and log samples from August 2003 to January 2004.
curve between water content and respiration proposed by Mielnick and Dugas (2000) because a similar regression curve was attained in our previous study using an automated chamber system (Jomura et al., 2005a). The following regression explained 61% of the variation in Rsnag and Rlog (Fig. 4): RCWD ¼ 17:614 e0:068T c ðupore þ 0:010Þð1:299 upore Þ2:023
(4)
4.3. Time trends in temperature and water content of the snags and logs Temperatures of snags, logs, air, and soil (Tsnag, Tlog, Tair, and Tsoil, the soil temperature at a 20-cm depth) during the measurement period ranged from 1.2 to 35.5 8C, 0.7 to 31.2 8C, 0.0 to 33.8 8C, and 3.0 to 26.5 8C, respectively (Fig. 5a). The temperatures of snags and logs showed diurnal changes that corresponded to changes in air temperature. The relationship between Tsnag and Tlog was obtained by linear regression
Fig. 3. Relationships (a) between chamber temperature (Tc) and respiration rate (RCWD) and (b) between water content (based on void volume, upore) and respiration rate, standardized at a temperature of 15 8C, for the snag and log samples.
M. Jomura et al. / Forest Ecology and Management 255 (2008) 149–155
Fig. 5. Half-hourly (a) temperature and (b) water contents of the snag and log samples. Symbols and range bars show the mean observed values and their standard deviations.
(Tsnag = 1.7677 + 1.0145 Tlog, r2 = 0.97). usnag and ulog varied with precipitation and drying and showed similar sawtooth patterns (Fig. 5b). However, usnag was about 20% of ulog based on a linear regression between the two parameters (usnag = 0.0353 + 0.1907ulog, r2 = 0.50). 4.4. Estimated annual Rsnag and Rlog We estimated Rsnag and Rlog using the functions based on temperature and water content described above (Eq. (4)) and the measured temperatures and water contents of the snags and logs. Rsnag and Rlog values ranged from 0.0 to 15.7 and 3.1 to 38.3 mg CO2 kg1 h1 (with corresponding means of 4.0 and 16.3 mg CO2 kg1 h1), respectively. Both respirations showed clear seasonal changes caused by temperature changes. Annual carbon losses from snags and logs were estimated to be 9.6 and 39.0 g C kg1 year1, respectively; thus, carbon losses from snags were roughly 25% of those from logs. As a result, the decay rate constants for snags and logs based on the single exponential decay model were 0.02 and 0.08 year1, respectively. 5. Discussion The relationship between RCWD and temperature was exponential (Fig. 3a), which was consistent with the results of many other respiration studies (e.g., Kirschbaum, 1995). The variation in RCWD at the same temperature was likely to be
153
caused by differences in water content. RCWD decreased at low and high water contents and reached a maximum at intermediate water contents (Fig. 3b). Two types of relationship between water content and RCWD have been reported in previous studies. Some studies reported that RCWD reached a maximum at an intermediate water content, as in our study (Yoneda, 1980; Jomura et al., 2005a), whereas others reported that RCWD did not decrease, even at high water contents (Chambers et al., 2001; Wang et al., 2002; Bond-Lamberty et al., 2003). This difference may result from the use of different units for water content. In the present study, we defined water content based on the void volume of the CWD (upore) whereas the studies that produced different results used water content based on the dry weight of CWD (udw). udw has a different maximum value, which depends on wood density. Even if the value of udw were the same, the actual quantity of water would depend on pore space. Boddy (1983) indicated that udw should not be used to compare udw between woods with different densities, but did not evaluate the effect of water on microbes. She proposed that measurement of the water in available airfilled pores in wood was effective. Yoneda (1980) categorized the relationships between RCWD and udw based on wood density and indicated that the maximum value of RCWD occurred at intermediate values of udw for each different density of CWD. Thus, if RCWD is modeled based on water content changes over a wide range of wood densities, we should use the water content based on the void volume of the CWD. The responses of Rsnag to changes in temperature and water content were similar to those of Rlog (Fig. 3), and the regression function explained 61% of the variation in Rsnag and Rlog (Fig. 4). This result suggests that microbial development in snags was not largely different from that in logs and that we could use the same function to estimate both respirations at our study site. The isolation of snags from the ground inhibits colonization by microbes, and this may be a primary factor responsible for the slower decomposition of snags (i.e., for the lower respiration rates). Our results showed that even snags could be decomposed if sufficient time had passed (about 30 years after the spread of pine wilt disease) for them to be invaded by microbes. The responses of Rsnag to two environmental factors (temperature and moisture content) are thus also useful for estimating CO2 efflux from CWD, including snags, at a landscape level. Respiration rates estimated using the function derived in this study (Eq. (4)) and intensive collection of environmental data for snags and logs allowed us to reproduce the observed data with our function (Fig. 6). This indicated that the difference in water content between the snags and logs was one of the main factors that controlled the differences between Rsnag and Rlog. However, as seen in the relationships between temperature, water content, and respiration, large variation in respiration data is observed even under the same environmental conditions (Fig. 3). This would reflect differences in microbial activity in the CWD. Microbial activity depends on the substrate chemical quality. Levels of nitrogen and recalcitrant materials such as lignin control the rate of CWD decomposition (e.g., Harmon et al., 1986; Berg and McClaugherty, 2003). We have previously tried to quantify
154
M. Jomura et al. / Forest Ecology and Management 255 (2008) 149–155
Fig. 6. Seasonal changes in estimated and observed Rlog and Rsnag and cumulative carbon loss from snags and logs. Estimates are based on the equation for respiration as a function of temperature and water content [RCWD ¼ 17:614 e0:068T CWD ðupore þ 0:010Þð1:299 upore Þ2:023 ] for the snag and log samples.
microbial biomass in CWD using a fumigation–extraction method to evaluate the relationship between microbial biomass and CWD respiration rate (Jomura et al., 2004). However, continuous measurements were impossible because destructive sampling is required in this method. Finding a non-destructive means of continuously observing and modeling microbial activity would permit more accurate estimates of the pattern of CWD respiration in the future. Annual rates of carbon loss from snags and logs were estimated to be 9.6 and 39.0 g C kg1 year1 (Fig. 6), resulting in decay rate constants of 0.019 and 0.081 year1, respectively. Jomura et al. (2005b) reported decay rate constants of 0.028 and 0.149 year1, respectively, for blocks of P. densiflora wood located above and at the surface of the forest floor at the same site. In this older experiment, weight loss due to fragmentation was excluded, so the decay rate constants expressed decomposition by mineralization (i.e., respiration loss) only. The decay rates of the block specimens were faster than those of the CWD respiration samples in the present study. This difference probably resulted from difference in the size of the sample of CWD; the block specimens were 3 cm on each side, whereas the mean size of the CWD samples in the present study was about 15 cm. According to Jomura et al. (2007), the respiration rate of 3-cm CWD would be about two times that of 15-cm CWD. Therefore, it is fair to say that the decay rate calculated using the annual respiration rate estimated from daily and seasonal changes in environmental factors almost fits the values calculated from the weight loss rate. The estimation of decay rates of organic material based on respiration measurements and function that incorporate the main factors that control decay is thus an appropriate approach for considering shortterm effects of measured environmental factors. 6. Conclusions We determined spatial variation in respiration from CWD (standing and downed) in a temperate secondary broad-leaved forest in Japan. The function that described this relationship by incorporating only two environmental factors (temperature and
moisture content) could explain 61% of the variation in both snag and log respiration rates, but a large amount of variation was still observed under the same environmental conditions. Thus, to determine the absolute value of CWD respiration rate, it will be necessary to incorporate both environmental factors and biological factors such as microbial activity and biomass. The estimation of decay rates of CWD, including snags, based on respiration measurements is an appropriate and effective approach for taking into account the effects of environmental factors and evaluating carbon budgets based on CO2 in forest ecosystems. References Asano, I., 1982. Wood Dictionary. Asakura Press, Tokyo (in Japanese). Berg, B., McClaugherty, C., 2003. Plant Litter. Springer-Verlag, Berlin. Boddy, L., 1983. Microclimate and moisture dynamics of wood decomposing in terrestrial ecosystems. Soil Biol. Biochem. 15, 149–157. Bond-Lamberty, B., Wang, C., Gower, S.T., 2003. Annual carbon flux from woody debris for a boreal black spruce fire chronosequence. J. Geophys. Res. 108 (D3), 8220. Busing, R.T., 2005. Tree mortality, canopy turnover, and woody detritus in old cove forests of the southern Appalachians. Ecology 86, 73–84. Chambers, J.Q., Schimel, J.P., Nobre, A.D., 2001. Respiration from coarse wood litter in central Amazon forests. Biogeochemistry 52, 115–131. Daniels, L.D., Dobry, K., Klinka, K., Feller, M.C., 1997. Determining year of death of logs and snags of Thuja plicate in southwestern coastal British Columbia. Can. J. For. Res. 27, 1132–1141. Fahey, T.J., 1983. Nutrient dynamics of aboveground detritus in lodgepole pine (Pinus contorta ssp. latifolia) ecosystems, southeastern Wyoming. Ecol. Monogr. 53, 51–72. Goto, Y., Kominami, Y., Miyama, T., Tamai, K., Kanazawa, Y., 2003. Aboveground biomass and net primary production of a broad-leaved secondary forest in the southern part of Kyoto prefecture, central Japan. Bull. For. For. Prod. Res. Inst. 387, 115–147. Gower, S.T., 2003. Patterns and mechanisms of the forest carbon cycle. Annu. Rev. Environ. Res. 28, 169–204. Hammond, H.E.J., Langor, D.W., Spence, J.R., 2001. Early colonization of Populus wood by saproxylic beetles (Coleoptera). Can. J. For. Res. 31, 1175–1183. Harmon, M.E., Franklin, J.F., Awanson, F.J., Sollins, P., Gregory, S.V., Lattin, J.D., Anderson, N.H., Cline, S.P., Aumen, N.G., Sedell, J.R., Lienkaemper, G.W., Cromack Jr., K., Cummins, K.W., 1986. Ecology of coarse woody debris in temperate ecosystems. Adv. Ecol. Res. 15, 133–302.
M. Jomura et al. / Forest Ecology and Management 255 (2008) 149–155 Hicks, W.T., Harmon, M.E., Myrold, D.D., 2003. Substrate controls on nitrogen fixation and respiration in woody debris from the Pacific Northwest USA. For. Ecol. Manage. 176, 25–35. Howard, E.A., Gower, S.T., Foley, J.A., Kucharik, C.J., 2004. Effects of logging on carbon dynamics of a jack pine forest in Saskatchewan, Canada. Global Change Biol. 10, 1267–1284. Janisch, J.E., Harmon, M.E., 2002. Successional changes in live and dead wood carbon stores: implications for net ecosystem productivity. Tree Physiol. 22, 77–89. Johnson, E.A., Greene, D.F., 1991. A method for studying dead bole dynamics in Pinus contorta var. latifolia–Picea engelmannii forests. J. Veg. Sci. 2, 523–530. Jomura, M., Kimura, H., Furusawa, H., Kominami, Y., Kanazawa, Y., 2004. The measurement of microbial biomass C in coarse woody debris by fumigation– extraction method. Memoirs of Graduate School of Science and Technology Kobe University 22B, 67–70 (in Japanese with English summary). Jomura, M., Kominami, Y., Kanazawa, Y., 2005a. Long-term measurements of the CO2 flux from coarse woody debris using an automated chamber system. J. Jpn. For. Sci. 87, 138–144 (in Japanese with English summary). Jomura, M., Takeda, F., Kominami, Y., Kanazawa, Y., Goto, Y., 2005b. Spatial variation in the decomposition rate of block specimens of Japanese red pine (Pinus densiflora Sieb. et Zucc.). Appl. For. Sci. 14, 83–88. Jomura, M., Kominami, Y., Tamai, K., Miyama, T., Goto, Y., Dannoura, M., Kanazawa, Y., 2007. The carbon budget of coarse woody debris in a temperate broad-leaved secondary forest in Japan. Tellus 59B, 211–222. Kirschbaum, M.U.F., 1995. The temperature dependence of soil organic matter decomposition, and the effect of global warming on soil organic C storage. Soil Biol. Biochem. 27, 753–760.
155
Krankina, O.N., Harmon, M.E., Kukuev, Y.A., Treyfeld, R.F., Kashpor, N.N., Kresnov, V.G., Skudin, V.M., Protasov, N.A., Yatskov, M., Spycher, G., Povarov, E.D., 2003. Coarse woody debris in forest regions of Russia. Can. J. For. Res. 32, 768–778. Laiho, R., Prescott, C.E., 2004. Decay and nutrient dynamics of coarse woody debris in northern coniferous forests: a synthesis. Can. J. For. Res. 34, 763– 777. Mackensen, J., Bauhus, J., Webber, E., 2003. Decomposition rates of coarse woody debris—a review with particular emphasis on Australian tree species. Aust. J. Bot. 51, 27–37. Mielnick, P.C., Dugas, W.A., 2000. Soil CO2 flux in a tallgrass prairie. Soil Biol. Biochem. 32, 221–228. Næsset, E., 1999. Decomposition rate constants of Picea abies logs in southeastern Norway. Can. J. For. Res. 29, 372–381. Olson, J.S., 1963. Energy storage and the balance of procedures and decomposers in ecological systems. Ecology 44, 322–331. Siitonen, J., Martikainen, P., Punttila, P., Rahh, J., 2000. Coarse woody debris and stand characteristics in mature managed and old-growth boreal mesic forests in southern Finland. For. Ecol. Manage. 128, 211–225. Wang, C., Bond-Lamberty, B., Gower, S.T., 2002. Environmental controls on carbon dioxide flux from black spruce coarse woody debris. Oecologia 132, 374–381. Yatskov, M., Harmon, M.E., Krankina, O.N., 2003. A chronosequence of wood decomposition in the boreal forests on Russia. Can. J. For. Res. 33, 1211– 1226. Yoneda, T., 1980. Studies on the rate of decay of wood litter on the forest floor. 3. Effect of moisture content on CO2 evolution from decaying wood. Jpn. J. Ecol. 30, 55–62.