Abstract
Knowledge of soil SO4 adsorption characteristics is required for predicting the fate of atmospheric S depositions on terrestrial ecosystems. This study developed two models for predicting the SO4 sorption characteristics of field moist soils from routinely measured soil physical and chemical properties. Sixty-two soil samples from the U.S. Northeastern (NE) and Southern Blue Ridge Province (SBRP) regions were used in model development. The first modeling approach used stepwise multiple regression to identify which combinations of soil properties best predicted each Langmuir constant. The Langmuir constant regression equations were then substituted back into the Langmuir model and used to predict SO4 sorption for each equilibrium SO4 concentration. This approach described SO4 sorption reasonably well for both regions (R2 of 0.57-0.89). In the second approach, a better fit of SO4 sorption was achieved (R2 of 0.81-0.92) using fewer soil properties by expressing the Langmuir constants as linear functions of selected soil properties and directly estimating the soil property coefficients by nonlinear regression. Soil properties used in the second model included: dithionite-citrate-extractable Al (Al(d)), total organic C (TOC), and extractable SO4 for the NE region; and Al(d), TOC, clay content, pH, and CEC for the SBRP region. These results indicated that regional soil SO4 sorption characteristics could be predicted for field moist soils with the knowledge of only a few routinely measured soil properties.
Original language | English (US) |
---|---|
Pages (from-to) | 426-432 |
Number of pages | 7 |
Journal | Journal of Environmental Quality |
Volume | 21 |
Issue number | 3 |
DOIs | |
State | Published - 1992 |
Externally published | Yes |
ASJC Scopus subject areas
- Environmental Engineering
- Water Science and Technology
- Waste Management and Disposal
- Pollution
- Management, Monitoring, Policy and Law