ICE Toolkit
The ICE tool requires the use of Javascript to enable it's many useful features. If you cannot enable
Javascript though, don't worry. You can produce the same results on your own with a little help.
The Download Model Data page allows users to download Microsoft Excel spreadsheets of
the model statistics for each available model. All variables needed to conduct model
calculations can be located there.
You will need to identify your surrogate species or predicted species/family/genus and use the tables
to find matching data to run your calculations. The predicted taxa toxicity can be calculated as:
y = 10^[ a + b*Log10(x)]
Where x is the toxicity value of the surrogate species, and a and b are the intercept and
slope, respectively. The lower and upper confidence bounds are calculated as:
Lower bound = 10^ {Log10(y) - t%CI*sqrt [MSE*1/(df+2)+(x*-x)2/Sxx ] }
Upper bound = 10^ {Log10(y) + t%CI*sqrt [MSE*1/(df+2)+(x*-x)2/Sxx ] }
Where y is the estimated toxicity value of predicted taxa, t%CI is the value of t at the
desired confidence level (90%, 95%, 99%), MSE is the mean square error, and Sxx is
the sum of squared deviations of the surrogate (Dowdy and Wearden 1991).
Please address all comments and questions to the webmaster Office of Research and Development |
Center for Environmental Measurement and Modeling |
Gulf Ecosystem Measurement and Modeling Division
This module is designed to mass process toxicity values predicted from species, genus, and/or family level models
from available surrogates. The output will provide predictions from all available models for the selected taxonomic
levels for up to 25 surrogates.
