For many real-world modeling applications, it is necessary to evaluate the model performance by comparing simulated outputs (e.g., heads) to measured field parameters (e.g., water levels in monitoring wells). MAGNET allows almost instant comparison of simulated heads with observed groundwater levels available across the model domain (steady-state, single time-step, or temporally-aggregated) or at discrete locations for all time-steps (time-series comparisons). This is made possible by realtime access to a vast amount of pre-processed water well records and monitoring well data embedded and live-linked to the MAGNET modeling platform
TASKS
A - Study the step by step short video tutorial under MAGNET Quick Tutorials: Model Calibration
B - Develop a MAGNET model that reproduces the simulation as depicted in the tutorial.
C - Briefly respond to the following questions:
- Explain the source of the "spread" in the water well data displayed in the 45 deg. plot shown above and in the tutorial
- Is it possible to reduce the spread in the water well data cloud, other than by taking more accurate measurements? Explain
- What other comparisons might be useful for evaluating model performance?