CHARGE FIFRA SAP Meeting - August 30, 2002
FIFRA SAP WEB SITE http://www.epa.gov/scipoly/sap/
Sheraton Crystal City Hotel
1800 Jefferson Davis Highway
Arlington, VA 22202
FIFRA SAP Meeting - August 30, 2002
Stochastic Human Exposure and Dose Simulation Model (SHEDS) : System Operation Review of a Scenario Specific Model (Sheds-Wood) to Estimate Children's Exposure and Dose to Wood Preservatives from Treated Playsets and Residential Decks Using EPA/ORD's SHEDS Probabilistic Model.
Documentation and Operation of SHEDS-Wood
Question 1 The User's Manual for SHEDS-Wood provides installation and operational instructions for the software. The Panel is requested to comment on the clarity and completeness of the User's Manual and the organization and user-friendliness of the model interface. Does the Panel have any suggestions for improving the User's Manual or the model interface?
Question 2 The Technical Manual for the SHEDS-Wood model provides an overview of the model construct and detailed descriptions of key model components. The Panel is requested to comment on the clarity, completeness and usefulness of this document with respect to describing the model construct and scientific principles underlying the model. Does the Panel have any suggestions for improving the Technical Manual?
Question 3 The Source Code Directory on the CD provides annotated code for the exposure and dose algorithms used in the SHEDS-Wood model. Are these algorithms consistent with the descriptions in the SHEDS-Wood Technical Manual?
Question 4 SHEDS is a probabilistic model that simulates exposure and dose for population cohorts and chemicals of interest. The model simulates individuals from the user-specified population cohorts by selecting daily sequential time-location-activity diaries from the EPA Consolidated Human Activity Database (CHAD). For this particular application of the SHEDS-Wood model, the user-specified population cohort chosen by OPP, is children 1 to 6 years of age who contact playsets and/or home decks. The Panel is requested to comment on whether this approach appropriately considers demographic factors and statistical representativeness for the scenario selected, given sample size considerations for each cohort. Can the Panel recommend alternative(s) approaches to simulate a population cohort for estimating wood preservative exposure to children?
Question 5 The SHEDS-Wood model simulates longitudinal activity patterns for individuals by constructing a 365-day profile using 8 CHAD diaries from the same age-gender cohort. These eight diaries consist of two from each of the four seasons, one sampled on a weekend and the other on a weekday. The Panel is requested to comment on the appropriateness of this approach, given sample size considerations and availability of longitudinal activity data. Does the approach provide a reasonable and realistic construct with respect to temporal variability in magnitude and frequency for children's exposure to wood preservatives from treated decks and playsets?
Question 6 The Panel is requested to comment on the Agency's selection of the (non-chemical specific) input values shown in Appendices 2 and 3 of the Technical Manual, especially the dermal transfer coefficient (TC) and days per year for outdoor playing, for which no data are currently available. Does the Panel have recommendations concerning the following:
A) Are there existing research (which the Agency may have missed) or recommended approaches or studies which could also be used for developing the input parameters for which few or no data are available (e.g., dermal transfer coefficient and days per year a child spends on/around treated playsets and home decks)?
B) Are the variability and uncertainty distributions assigned to these non-chemical specific input parameters appropriate?
C) Is the bootstrap approach [Frey et al. (2002)] for fitting uncertainty distributions appropriate or are there alternative approaches which are preferable?
Model Results and Applications
Question 7 The Panel is requested to comment on the statistical
diagnostic tools used by SHEDS for analyzing model results (e.g.,
variability analyses, sensitivity analyses, uncertainty analyses)
and on the model capabilities for displaying results (e.g., summary
statistics tables, pie charts, CDFs). Are there additional analyses
or outputs that would be useful?
Question 8 Does the Panel recommend any additional refinements or modifications to the model (e.g., equations, assumptions, or algorithms) in order to make it more relevant to this particular application?