Soil dissolved organic carbon in terrestrial ecosystems: global budget, spatial distribution and controls

Aims: Soil dissolved organic carbon (DOC) is a primary form of labile carbon in terrestrial ecosystems and therefore plays a vital role in soil carbon cycling. This study aims to quantify the budgets of soil DOC at biome- and global levels and to examine the variations in soil DOC and their environmental controls. Location: Global Time period: 1981 - 2019 Method: We compiled a global dataset and analyzed the concentration and distribution of DOC across 10 biomes. Results: Large variations in DOC are found among biomes across space and the soil DOC concentration declines exponentially along soil depths. Tundra has the highest soil DOC concentration in 0 - 30 cm soils (453.75 (95% confidence interval: 324.95 – 633.5) mg·kg-1); whereas tropical and temperate forests have relatively lower DOC concentrations, ranging from 30.20 (24.78 - 36.80) mg·kg-1 to 54.54 (49.77 – 59.77) mg·kg-1. DOC generally accounts for < 1% of total organic carbon in soils, and DOC in 0 - 30 cm contributes more than half of total DOC in 0 - 100 cm soil profile. Furthermore, variations in DOC are primarily controlled by soil texture, moisture, and total organic carbon. Main conclusion: A global synthesis is combined with an empirical model to extrapolate the DOC concentration along soil profiles across the globe, and global budgets of DOC are estimated as 7.20 Pg C in top 0 - 30 cm and 12.97 Pg C in 0 - 100 cm, respectively, with a considerable variation among biomes. The strong soil texture control but weak TOC control on DOC variations suggest that the investigation of physical protection of soil organic carbon might need to expand to consider the labile C in soils. The global maps of DOC concentration serve as a benchmark for validating land surface models in estimating carbon storage in soils.The database represents 171 sites from 33 countries, of which 34.1% in Europe, 34.6% in Asia, 27.2% in America and < 4% in other continents. These data points represent soil samples from various depths of 0-100 cm. Available auxiliary information of the sampling sites was also retrieved, including the latitude and longitude, mean annual air temperature (MAT), mean annual precipitation (MAP), soil pH, TOC, total nitrogen concentration (TN), vegetation type, soil texture type, and sampling date. The database used in this study was organized in April 2019 and was updated and finalized on November 20th, 2019. Funding provided by: National Natural Science Foundation of ChinaCrossref Funder Registry ID: http://dx.doi.org/10.13039/501100001809Award Number: 41730643, 41975150, 41701198Funding provided by: Basic Scientific Research Operating Expenses of Heilongjiang Provincial Universities*Crossref Funder Registry ID: Award Number: 2018CX05Funding provided by: San Diego State UniversityCrossref Funder Registry ID: http://dx.doi.org/10.13039/100007099Award Number:The data for DOC concentrations were collected from the publications by searching "soil DOC" in Google Scholar and the Web of Science. The data points were derived from tables containing soil DOC or extracted by the Engauge Digitizer software (Version 4.1) from figures in collected publications. The data points with reported soil DOC concentrations greater than total organic C (TOC) concentration were excluded from the database. In total, 3869 data points were retrieved from 107 papers published during 1981 - 2019