scholarly journals Spatial Modeling of Soil Lime Requirements with Uncertainty Assessment Using Geostatistical Sequential Indicator Simulation

2017 ◽  
Vol 07 (07) ◽  
pp. 133-148 ◽  
Author(s):  
Jussara de Oliveira Ortiz ◽  
Carlos Alberto Felgueiras ◽  
Eduardo Celso Gerbi Camargo ◽  
Camilo Daleles Rennó ◽  
Manoel Jimenez Ortiz
Keyword(s):  
Spatial Modeling ◽  
Uncertainty Assessment ◽  
Sequential Indicator Simulation ◽  
Indicator Simulation

Assessing spatial distribution and joint uncertainty of TPH-fractions: Indicator kriging and sequential indicator simulation

2007 ◽  
Vol 87 (5) ◽  
pp. 551-563
Author(s):  
Carol Luca ◽  
Bing C Si ◽  
Richard E Farrell
Keyword(s):  
Spatial Distribution ◽  
Petroleum Hydrocarbon ◽  
Indicator Kriging ◽  
Hydrocarbon Contamination ◽  
Contaminated Sites ◽  
Sequential Indicator Simulation ◽  
Maximum Probability ◽  
Dependent Case ◽  
Indicator Simulation ◽  
Independent Case

Petroleum hydrocarbon (PHC) contamination is one of the most common contaminants in soils and remediation of PHC-contaminated sites requires methods for characterizing the spatial distribution of PHC on a site. Few studies have compared the performance of indicator kriging (IK) and sequential indicator simulation (SIS) in site characterization of petroleum-contaminated sites, or the application of these methods given the fraction based guidelines. The objectives of this study were to determine if IK and SIS indicate similar contaminated areas and to examine how the probability of exceeding thresholds changes when multiple fractions are considered simultaneously. An abandoned refinery near Kamsack, Saskatchewan, characterized by clay-textured soils was sampled and analyzed for PHC fractions (F2 and F3). The probability of a location exceeding a fraction’s remediation criteria was determined using IK and SIS. Based on critical probability thresholds, IK indicated a greater area was contaminated by F2 (6.3%) and F3 (0.8%) than SIS (4.5 and 0.6%, respectively). When the remediation criteria for both F2 and F3 were considered simultaneously, “dependent” and “independent” cases were examined. The dependent case assumed perfect correlation and used the maximum probability of either F2 or F3 as the new estimate. The independent case assumed no correlation and evaluated the probability of F2 > 2500 mg kg–1 or F3 > 6600 mg kg–1. The dependent case resulted in a smaller contaminated area than the independent case in both IK and SIS. On this site the differences between the two methods were small, although IK did smooth the distribution. Key words: Sequential indicator simulation, indicator kriging, geostatics, petroleum hydrocarbon contamination, uncertainty

Sign in / Sign up

Export Citation Format

Share Document