PROJECT GIGALOPOLIS

 

Step 4: Selecting Coefficient Ranges > Selecting coefficients with Optimum SLEUTH Metric (OSM)

Optimum SLEUTH Metric (Dietzel and Clarke, 2007) code is provided in the Download page. After each phase of calibration OSM code can be run using the control_stats.log file to find out the 'top 50' best fit values.

1. Download the OSM code, unzip and place it in the 'Output' directory.

2. Execute the OSM run

     prompt% ./readdata2

     autoprompt 'Enter the input file name:' % control_stats.log

3. This generates 'top50.txt' which stores the top 50 best fit set of coefficient values.

4. For the top three rankings: take high and low values of each of the coefficients.It is possible for more than one run to have the same score, creating a tie

5. For each coefficient: in the scenario file to be used for fine calibration, low values are set to _START

6. For each coefficient: in the scenario file to be used for fine calibration, high values are set to _STOP

7. A _STEP value is selected that will increment between the _START and _STOP values 4-6 times
If only one coefficient value sorts into the top 10 (e.g.; "1" for dispersion and spread) select _START, _STEP, and _STOP values that will explore a finer coefficient space around the value

 

Top 5 scores from an example study coarse calibration, sorting only on the OSM:

osm1

Coefficient range for fine calibration:

coefficient type {_START - _STOP, _STEP}
dispersion {70 - 100, 8}
breed {1 - 1, 1}
spread {75 - 83, 5}
slope {1 - 24, 5}
road gravity {25 - 85, 10}


Top 5 scores from an example study final calibration, sorting only on the OSM:

osm

 

Coefficient values used to predict growth:

coefficient type PREDICTION_*_BEST_FIT
dispersion 100
breed 1
spread 75
slope 24
road gravity 1

 

What is a good OSM value?

 

 

 

 

 

back to Implement

USGSUCSB