Abstract

Studies on the fate of toxic chemicals in soils are often reported with a minimum of descriptive statistics. Use of modeling techniques to describe the kinetics of chemical degradation provides a better understanding of the fate of chemicals in soil systems. When modeling nonlinear systems, assumptions made about the error term greatly influence the parameter estimation. Inappropriate use of linearization and failure to account for autocorrelated errors may result in inaccurate models. Information is also needed about the effects of the magnitude of autocorrelation on parameter estimation. The exponential decay function was chosen to fit the data obtained from a TNT (2, 4, 6-trinitrotoluene) degradation experiment in soil using four different error assumptions. Estimates of the rate constant (k) and other parameter estimates changed appreciably as assumptions about the error term changed. Simulation studies indicated that modeling data from chemical decomposition studies with an independent error assumption resulted in unreliable k estimates when the autocorrelation was large. A two-step procedure was used to fit an exponential autocorrelated (AR(1)) model. Overall, the exponential function with the additive-correlated error assumption provided the best fit for TNT degradation data. In essence, the kinetic rate constant obtained through model fitting in chemical decomposition studies provides a great deal of useful information to scientists. However, the researcher must be aware of the fact that making correct assumptions about the error term is extremely critical for obtaining accurate and precise estimates of k.

Creative Commons License


This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Share

COinS
 
Apr 25th, 9:00 AM

MODELING THE FATE OF TOXIC CHEMICALS IN SOILS

Studies on the fate of toxic chemicals in soils are often reported with a minimum of descriptive statistics. Use of modeling techniques to describe the kinetics of chemical degradation provides a better understanding of the fate of chemicals in soil systems. When modeling nonlinear systems, assumptions made about the error term greatly influence the parameter estimation. Inappropriate use of linearization and failure to account for autocorrelated errors may result in inaccurate models. Information is also needed about the effects of the magnitude of autocorrelation on parameter estimation. The exponential decay function was chosen to fit the data obtained from a TNT (2, 4, 6-trinitrotoluene) degradation experiment in soil using four different error assumptions. Estimates of the rate constant (k) and other parameter estimates changed appreciably as assumptions about the error term changed. Simulation studies indicated that modeling data from chemical decomposition studies with an independent error assumption resulted in unreliable k estimates when the autocorrelation was large. A two-step procedure was used to fit an exponential autocorrelated (AR(1)) model. Overall, the exponential function with the additive-correlated error assumption provided the best fit for TNT degradation data. In essence, the kinetic rate constant obtained through model fitting in chemical decomposition studies provides a great deal of useful information to scientists. However, the researcher must be aware of the fact that making correct assumptions about the error term is extremely critical for obtaining accurate and precise estimates of k.