SHORT COMMUNICATION J For Res (2005) 10:57–60 © The Japanese Forest Society and Springer-Verlag Tokyo 2005 DOI 10.1007/s10310-004-0102-2 Wen Jie Wang · Yuan Gang Zu · Hui Mei Wang Takashi Hirano · Kentaro Takagi · Kaichiro Sasa Takayoshi Koike Effect of collar insertion on soil respiration in a larch forest measured with a LI-6400 soil CO 2 flux system Received: January 30, 2004 / Accepted: April 27, 2004 W.J. Wang · Y.G. Zu · H.M. Wang Key Laboratory of Forest Plant Ecology, Northeast Forestry University, Harbin, PR China T. Hirano Graduate School of Agriculture, Hokkaido University, Sapporo, Japan K. Takagi · K. Sasa · T. Koike (*) Hokkaido University Forests, FSC, Hokkaido University, Sapporo 060-0809, Japan Tel. +81-11-706-3854; Fax +81-11-706-3450 e-mail: tkoike@exfor.agr.hokudai.ac.jp Abstract Little information is available on the effect of root cutting by the collar pre-insertion technique on soil respira- tion. In this study, we found that soil respiration rates de- creased with increasing depth of collar insertion in both the “with live roots intact” and “with live roots severed” treat- ments, but the rate of decrease was substantially higher in the former. The cutting of roots, especially fine roots, may be responsible for this result. Key words Larch (Larix gmelinii) · Soil respiration rate · Collar insertion · Root biomass Introduction Soil respiration rate can now be made routinely with a LI-6400 soil CO 2 flux system (LiCor, Lincoln, NE, USA). The soil chamber of the system is vented to equilibrate the pressure inside and outside the chamber, since pressure differences can have a strong effect on estimates of soil respiration (Nakayama 1990; Norman et al. 1992; LiCor 1997). Further improvements in soil respiration measure- ments can be achieved by the insertion of thin-walled polyvinyl chloride (PVC) collars prior to taking the mea- surements (Norman et al. 1992). These collars minimize soil surface disturbance and reduce the sudden flushing of CO 2 in the soil and litter. Despite these improvements, there are still large variations in soil respiration even within a single forest, due to biotic and abiotic factors (Lloyd and Taylor 1994; Fang et al. 1998; Davidson et al. 1998; Buchmann 2000; Fang and Moncrieff 2001). One of these factors is the contribution of root respiration, which can be 40%–60% of the total soil respiration rate (Bowden et al. 1993; Hanson et al. 2000). It is, therefore, critical to determine the effect of inserting the root-cutting collar on soil respiration. How- ever, available information on the effect of the collars on soil respiration measurement is still limited (Hanson et al. 2000). Ignoring the effect of the root-cutting collar can therefore result in misleading soil respiration rates. The problem is expected to be more serious in forests with a shallow root system, such as those dominated by Larix gmelinii. These forests are widespread in central Siberia, Russia (Kajimoto et al. 2003) and northeast China (Wang et al. 2002). To examine the effect of inserting the root-cutting collar on soil respiration, a long-term CO 2 flux site in a L. gmelinii forest in northeast China was selected as the study site (Wang et al. 2002). Since soil trenching is a method commonly used for estimating root respiration (Hanson et al. 2000; Lee et al. 2003), we also investigated the effect of collar insertion in such a soil trench. Study site and methods Study site Measurements were made in a larch forest (L. gmelinii (Rupr.) Rupr.) 36 years old in 2003, located in the Laoshan Experimental Forest of the Northeast Forestry University in China (45°20'N, 127°34'E). The mean annual precipita- tion at Laoshan Experimental Forest was 723 mm and the annual mean air temperature was 2.8°C. The density of the larch stand was 1150 stems ha -1 . The mean height was 18 m and mean diameter at breast height (DBH) was 17.2 cm. The soil here is characterized as a dark-brown forest soil and with about 10% soil organic matter, and pH values of 5.3–6.0. Depth of the litter layer at our study plot was about 2 cm. A more detailed description of this research site is given by Shi et al. (2001).