KayShelton

UPDATE: I have managed to establish my colleague actually only used post 1981 data for their evaluation (I don't have the .ini files unfortunately). I have re-run TCRM V3 using...

Further investigations: I have determined that the data generated from the historical dataset during the DataProcessing in V3 is identical to that from using V2. The all_cell_cdf files are different,...

So I did further investigating, and the choice of KDE doesn't appear to make a great deal of difference except to the initiation day. Comparison plot for the genesis pressure:...

There is not overly much spatial variation, but that is primarily due to the small sample size across the domain. With only about 115 TCs across the entire domain in...

Thinking again about the poor representation of the historical minimum pressure distribution by the synthetic dataset, in particular the fat tail at the lower end of the distribution. I think...

I have tried a possible alternative for the minCP threshold: `# if (pressure[i] < (pstat.min[cellNum] - # 4. * pstat.sig[cellNum])): ##### added to allow some greater variability ##### sets threshold...

So, interestingly, resolving the bug in https://github.com/GeoscienceAustralia/tcrm/issues/87 does appear to have also fixed the slight bias in the minCP histogram toward higher than observed pressures. Below is the results for...

I have been following up on the low track densities which are evident in the evaluation simulations for the South Pacific, but potentially more problematic in regions with very few...

For a limited test of 5 simulations of 59 years for the South Pacific, applying the 4 conditions above to the TrackGenerator module results in a significant improvement in TC...

Musing on the poor central pressure distributions, I wonder if the step pressure change formulation is entirely appropriate. Using the logistic distribution as the population distribution to sample from is...