Belowground primary production by carbon isotope decay and longterm root biomass dynamics

The isotope decay method of estimating below-ground net primary production (BNPP) has the potential to overcome the assumptions and biases associated with traditional methods. Isotope loss through in situ decomposition after pulse-labeling is considered the inverse of production, and turnover times are estimated by regression to time of zero remaining isotope. Method development and estimates of production were previously published using 4 years of data, which showed a clear linear loss rate over time. A slow, distinctly different phase in isotope loss developed 5-10 years postlabeling. We assess reasons for the two-phase loss functions and the implications for estimates of BNPP and compare the isotope decay method with standard coring methods over a 13-year period. Reasons for the two-phase dynamics of carbon 14 (C-14) loss could include various biological and/or methodological factors. Results suggest that C-14 in soil embedded in roots as they grow, a small proportion of roots that live for a much longer time than the majority of roots, and method of separating roots from soil organic matter may influence estimates of BNPP by isotope methods. Remobilization of label in structural tissue or reuptake of label from the soil did not appear to be responsible for the slow, second phase of loss dynamics. Isotope decay produced more reliable estimates than standard coring methods. Estimates using harvest sum of increments were zero in 6 of 13 years. Thirteen years of root biomass data showed no predictable trend over winter or consistent seasonal pattern, although longer-term cycles were evident. Aboveground:belowground ratios were generally smaller during dry periods, but root biomass was not as responsive as aboveground biomass to annual precipitation.