Quantifying Tacit Knowledge about Soil Organic Carbon Stocks Using Soil Taxa and Official Soil Series Descriptions. Issue 5 (20th September 2013)

• Record Type: Journal Article
• Authors: Wills, Skye; Seybold, Cathy; Chiaretti, Joe; Sequeira, Cleiton; West, Larry
• Is Part Of: Soil Science Society of America Journal. Volume 77:Issue 5(2013)
• Page Start: 1711
• Page End: 1723
• Publication Date: 2013-09-20
• Subjects: Soils -- United States -- Periodicals; Soil science -- Periodicals; Periodicals; 631.4973
• DOI: 10.2136/sssaj2012.0168
• Languages: English
• ISSNs: 0361-5995

Abstract

Abstract : Previous assessments of soil organic carbon (SOC) stocks have used soil maps and other tacit knowledge from soil survey to aggregate and extrapolate data without assessing the relationship between SOC and the tacit information used. The objective of this paper is to outline a process that translates tacit class‐based knowledge into semiquantitative information about SOC stocks. This information was used to create a soil‐based strata for sample selection to inventory SOC stocks. The process was completed in three steps: (i) official series descriptions (OSDs) were used to compile comprehensive soil information, (ii) an ordinal scoring system was developed and applied to link soil properties to likely SOC stocks, and (iii) hierarchical clustering was applied to cluster soils into groups for sampling. The scores and groups were tested using pedon SOC stocks from the National Cooperative Soil Survey laboratory database. Expert soil scientists interpreted individual terms used in Soil Taxonomy and OSDs for their explicit or implicit relationships with SOC stocks. Taxa, soil moisture regime, and temperature regime were separated as individual factors. Other information from the OSDs including family particle size class, depth to restrictive layer, and drainage class were used to translate qualitative information into ordinal scores. Classes for taxonomic order, soil moisture regimes, and drainage class were significantly different in pedon SOC stocks across and within regions. Scores within each factor (taxonomy, moisture regime, drainage class, temperature regime, and soil depth) were also significantly different from one another in pedon SOC stocks. Ordinal scores were used to cluster soils within each of 17 regions. Soil groups were significantly different from one another in SOC stocks in each of the 17 sampling regions. The process was considered satisfactory and was used to stratify soils for SOC stock sampling in the conterminous United States. A similar process could be applied to any soil property over any area of interest to evaluate tacit knowledge for communicating soil information, grouping soils for interpretation, or creating strata for distributing samples.