Methodologies based on 13 C-enrichment (E), 13 C-depletion (D) and 13 C-natural abundance (NA) to estimate the stability of biochar in soil were critically examined. The stability of 13 C-enriched biochar can be estimated by the quantitative recovery of excess 13 C, either in the soil or in evolved CO 2 . Both approaches have advantages and disadvantages. Recovery in the soil is a measure of both residual biochar 13 C + 13 C immobilised in soil organic matter during biochar decomposition. Variable proportions of organic-and inorganic-C are present in alkaline biochars, and few data exist on the uniformity of labelling, which is a basic requirement of the respired 13 CO 2 and E methodology.The E technique has had limited application due to the cost and difficulty of obtaining a uniformly-enriched feedstock through continuous labelling of plants with 13 CO 2 at a constant 13 C enrichment. In contrast, the NA technique has been widely applied. The NA and D techniques are in situ methods that involve the addition of C 4 -derived biochar to a C 3 -soil or vice versa. Stability is estimated by a two-end-member mixing model that allows the proportion of evolved CO 2 derived from the biochar (Cdfb) to be estimated. The mixing model has recently been misused to estimate the Cdfb of 13 C-enriched biochar, with 13 C-abundance expressed as erroneously large δ values. 13 C-based methods provide a yardstick against which rapid stability tests should be evaluated. While numerous laboratory incubation comparisons have been conducted, very few field-based data have been published.
Highlights• 13 C methods for estimating biochar stability are based on uniform isotopic labelling.• Organic and inorganic constituents of biochar may not be uniformly labelled.• Expression of 13 C enriched biochar as large δ values (>500 units) rather than atom fraction excess led to larger errors in stability estimation.• Few 13 C field-based estimates of biochar stability exist.
K E Y W O R D S13 C, 13 CO 2 , biochar stability, two-end-member model, uniform labelling, δ 13 C Phillip Chalk and Christopher J. Smith contributed equally to this study.