SoilHeterogeneity of C and 13C

Bohm, S., G.P. Robertson, and E.A. Paul

Presented at the ASM at Snowbird (2000-08-02 to 2017-12-05 )

The presence of short-range autocorrelation in soil nutrients has been demonstrated in a number of studies (Kluitenberg et al., 1997; Palmer, 1990; Schlesinger et al., 1996). The ranges of autocorrelation reported in different studies are dependent on the sample size, and sample intervals. The range of autocorrelation reported varies between 80 m for total organic N in an Iowa farm field (Cambardella et al., 1994), to 20 m for nitrate and ammonium in an old field community in Michigan (Robertson et al., 1997) to less than 2 m for nitrate in a southern Quebec forest ecosystem (Lechowicz and Bell, 1991).The presence of short-range autocorrelation in soil nutrients has been demonstrated in a number of studies (Kluitenberg et al., 1997; Palmer, 1990; Schlesinger et al., 1996). The ranges of autocorrelation reported in different studies are dependent on the sample size, and sample intervals. The range of autocorrelation reported varies between 80 m for total organic N in an Iowa farm field (Cambardella et al., 1994), to 20 m for nitrate and ammonium in an old field community in Michigan (Robertson et al., 1997) to less than 2 m for nitrate in a southern Quebec forest ecosystem (Lechowicz and Bell, 1991).Total C was reported to be auto-correlated at distances of less 7.4 m in an uncultivated successional field at KBS (Robertson et al., 1993). This is more than 10 times the range observed in the poplar plots of this study, however, samples in the first lag class were separated by 1.2 meters. When samples were taken at 5 m intervals, in an Iowa field, total C was auto-correlated to 109 – 129 m (Cambardella et al., 1994). Both studies used compositing to reduce the small-scale variability. Sampling at meter-scale distances will not reveal sub-meter heterogeneity. Stoyan et.al. (2000) measured the small-scale heterogeneity to reflect the actual microbial and plant root respiration. This resulted in large nugget variances for the variates even at lag distances of less than 1 meter. To understand the carbon cycling in the soil it is important to understand the distribution of carbon in the soil. We investigated the small scale variability of C, N and 13C in nine year old poplar plots during 1999.Total C was reported to be auto-correlated at distances of less 7.4 m in an uncultivated successional field at KBS (Robertson et al., 1993). This is more than 10 times the range observed in the poplar plots of this study, however, samples in the first lag class were separated by 1.2 meters. When samples were taken at 5 m intervals, in an Iowa field, total C was auto-correlated to 109 – 129 m (Cambardella et al., 1994). Both studies used compositing to reduce the small-scale variability. Sampling at meter-scale distances will not reveal sub-meter heterogeneity. Stoyan et.al. (2000) measured the small-scale heterogeneity to reflect the actual microbial and plant root respiration. This resulted in large nugget variances for the variates even at lag distances of less than 1 meter. To understand the carbon cycling in the soil it is important to understand the distribution of carbon in the soil. We investigated the small scale variability of C, N and 13C in nine year old poplar plots during 1999.

Back to meeting | Show |
Sign In