Variable carbon recovery of Walkley-Black analysis and implications for national soil organic carbon accounting

There is considerable interest in the computation of national and regional soil carbon stocks, largely as the result of the provisions of the Kyoto Protocol. Such stocks are often calculated and compared without proper reference to the uncertainties induced by different analytical methodologies. We illustrate the nature and magnitude of these uncertainties with the present soil organic carbon (SOC) study in Belgium. The SOC recovery of the Walkley-Black method was investigated based on a database of 475 samples of silt loam and sandy soils, which cover different soil depths and vegetation types in northern Belgium. The organic carbon content of the soil samples was measured by the original Walkley-Black method and by a total organic carbon analyser. The recovery was computed as the ratio of these two results per soil sample. Land use, texture and soil sampling depth had a significant influence on the recovery as well as their three-way interaction term (land use x texture x sampling depth). The impact of a land use, texture and sampling depth dependent Walkley-Black correction on the year 2000 SOC inventory of Belgium was determined by regression analysis. Based on new correction factors, the national SOC stocks increased by 22% for the whole country, ranging from 18% for cropland to 31% for mixed forest relative to the standard corrected SOC inventory. The new recovery values influenced therefore not only C stocks in the year 2000, but also the expected SOC change following land use change. Adequate correction of Walkley-Black measurements is therefore crucial for the absolute and comparative SOC assessments that are required for Kyoto reporting and must be computed to take into account the regional status of soil and land use. 'Universal' corrections are probably an unrealistic expectation.