Exercise #13 |
||||
 Previous |
|
 Next |
||
- Problem #1 - Using the nine output files from the previous section, determine the time variation of the emission rate using the ratio method.
- Hint - Instead of using a simple ratio of Measured/Calculated to determine the release rate each hour, use the values already tabulated in the coefficient matrix (c2array.csv) to compute a regression of M/C using the Excel slope function.
- Solution - The hourly release rate Excel solution shown on line 22 also shows considerable over-prediction.
- Problem #2 - We've found that the SVD or Cost function solutions are very sensitive to model error. We could create an ensemble mean to reduce error but this may take a more advanced level of script programming. In the last section, exercise #12, we found from the physics ensemble, that the simulation using the Beljaars turbulence (kblt=1) provided the best results. Recreate the nine output files for each emission time using this approach and compute the emission rate using SVD and cost function solutions.
- Hint - The simplest approach is to edit the script with the additional namelist variables.
- Solution - Solving the TCM shows that although the large over-predictions during the emission times are less, overall the SVD solution shows over-prediction by a factor-of-four compared to the original result. The cost function solution with the optimized dispersion is comparable to the original result.