Premium
Spatial Variability of Hydraulic Properties in the Emporia Series
Author(s) -
Albrecht K. A.,
Logsdon S. D.,
Parker J. C.,
Baker J. C.
Publication year - 1985
Publication title -
soil science society of america journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.836
H-Index - 168
eISSN - 1435-0661
pISSN - 0361-5995
DOI - 10.2136/sssaj1985.03615995004900060033x
Subject(s) - replicate , sampling (signal processing) , series (stratigraphy) , spatial variability , mathematics , hydrology (agriculture) , coefficient of variation , soil science , variance (accounting) , environmental science , statistics , geology , geotechnical engineering , business , computer vision , paleontology , accounting , filter (signal processing) , computer science
Variability of soil hydraulic properties of one soil series is evaluated, observation scale effects are analyzed, and sampling requirements for series characterization are determined. One site was selected in each of four counties. Every site had four subsites about 30‐m apart. Undisturbed cores were taken in triplicate about 0.3‐m apart at each subsite and disturbed samples taken in single replicate within sampling pits at each subsite. Three depths were examined—the A, E, and Bt horizons. Coefficients of variation (CV) for moisture retention data were low ranging from 1.0 to 15.4%, while CVs of saturated hydraulic conductivities ( K s ) ranged from 112 to 297%. Nested analyses of variance revealed that variance generally increased with increasing sample separation distance. The majority of the variance occurred at the site level. Numbers of samples needed to estimate means to selected confidence levels for series characterization were determined for each parameter by depth. Assuming costs to be equal between sites, subsites, and replicates, it was found that 10 to 26 sites with one subsite per site and one replicate per subsite should be sampled to adequately describe moisture retention to within 0.05 m 3 m −3 and 7 to 15 sites would be necessary for K s evaluation to within one order magnitude. Results emphasize the need to sample many sites for hydraulic properties which have most of their variance at this level.