B43I-0669
Differences in pedotransfer functions of bulk density lead to high uncertainty in soil organic carbon estimation at regional scales: Evidence from Chinese terrestrial ecosystems
Thursday, 17 December 2015
Poster Hall (Moscone South)
L. Xu1, Guirui Yu2, Nianpeng He2, Ding Wen2, Honglin He1 and Yang Gao2, (1)Institute of Geographic Sciences and Natural Resources Research, Chinese Academy of Sciences, Beijing, China, (2)IGSNRR Institute of Geographic Sciences and Natural Resources Research, CAS, Beijing, China
Abstract:
Soil contains the largest pool of organic carbon in terrestrial ecosystems. Accurate estimation of soil organic carbon (SOC) storage is important for evaluating carbon sequestration of terrestrial ecosystems at regional scale. How the selected pedotransfer functions (PTFs) of bulk density (BD) influence the estimates of SOC storage is still unclear at large scales, although BD is an important parameter in all equations. Here, we used data from the second national soil survey in China (8210 soil profiles) to evaluate the influence of eight selected PTFs on the estimation of SOC storage. The results showed that different PTFs may result in a higher uncertainty of SOC storage estimation, and the coefficient of variation (CV, %) for the eight PTFs varied from 10.61% to 70.46% (mean = 12.75%). The observed CV values were higher in the 0–20 cm layer (12.48%) than in the 20–100 cm layer (10.05%). CV values were relatively stable (10–15%) when SOC content ranged from 0.13% to 3.45%. The findings indicate that PTFs may be used cautiously in soils with higher or lower SOC content. Estimates of SOC storage in the 0–100 cm soil layer varied from 67.19 to 95.97 Pg C in the eight PTFs in China, with an average of 87.36 ± 8.93 Pg C (CV = 10.23%). Our findings provide the insight that differences in PTFs are important sources of uncertainty in SOC estimates, except for the different SOC data sets and estimation methods. The development of better PFTs, or the integration of various PFTs, is essential to accurately estimate SOC storage at regional scales.