Calibration (Model vs Data)
EE has been designed to quickly generate report quality plots of model to data comparisons and model results statistics. EE provides the tools to link the appropriate model cells/layers to each specific data set. Once these links are made, they are saved in the model configuration files for use in any subsequent model run. Each linkage can be disabled and/or enabled to report only what is needed for the current calibration stage. Each of these tools can be manually run at any time or EE can be configured to automatically generate these plots and statistics at the end of the run.
Time Series Plots
Time Series comparison plots can be configured to link observed data to the predicted model values. The example here shows temperature data compared simulated temperature modeled with EFDC+. Time Series comparison plots can be configured to link observed data to the predicted model values.
A wide range of statistical metrics can be automatically generated by EE and provided in summary format for each station configured. Some of the statistical measures include: Average Error, Relative Error, Average Absolute Error, Root Mean Square Error, and the Nash-Sutcliffe coefficient. EE outputs these in such a way that they can be easily imported into Excel or Word for reporting purposes.
For large water bodies, such as the San Francisco Bay and Delta, data collection is often routinely collected from a single moving boat. This produces a data set that is spatially and temporally variable for a single “Cruise”. EE allows the visual comparison of these data to the model results by defining a “cruise line” which corresponds to a transect of data sampled on a waterbody. EE extracts the data from the user-defined line at the specific time of the data collection at that point and compares this with measured data.
Vertical Profile Plots
Vertical profile plots are essential in developing calibrated models where there is significant stratification in the water column. EE shows the comparison between model and data for multiple model-data vertical profiles on each page.
EE plots correlations between the model and measured data, optionally displaying R-Squared and other statistics on each plot. An example of model data correlation plot comparison for water surface elevation is shown below. Note that observed data are shown on the x-axis and the model results are shown on the y-axis.
The flux comparison tool is provided to assist the user in calibration of flow or constituent flux across a user-defined cross section. Any number of cross sections can be used to generate time series comparisons between the measured and modeled flows. The flux comparison tool is provided to assist the user in calibration of flow or constituent flux across a user-defined cross section. Any number of cross sections can be used to generate time series comparisons between the measured and modeled flows.