This week I found a few issues with the biogeochemistry inputs which could have been respondible for nans in the output file. For example, I had a line of code that replaced zeros with nans, and I found some rivers that introduced negative TIC values to the system. After fixing these issues, the biogeochemistry output files looked complete. Unfortunately, fixing the biogeochemisty did not correct the Birch Bay WWTP blow up issue. I tested moving Birch Bay WWTP seaward by one grid cell. This model run did not crash, however the velocities near the WWTP still look unstable. Nans Ecology padded state variables with zeros for WWTPs that opened midway through the the timeseries. For example, a WWTP that opened in 2005 woul
TRAPS Debugging Part III | Weekly Research Update
This week I found a few issues with the biogeochemistry inputs which could have been respondible for nans in the output file. For example, I had a line of code that replaced zeros with nans, and I found some rivers that introduced negative TIC values to the system. After fixing these issues, the biogeochemistry output files looked complete. Unfortunately, fixing the biogeochemisty did not correct the Birch Bay WWTP blow up issue. I tested moving Birch Bay WWTP seaward by one grid cell. This model run did not crash, however the velocities near the WWTP still look unstable. Nans Ecology padded state variables with zeros for WWTPs that opened midway through the the timeseries. For example, a WWTP that opened in 2005 woul
https://ajleeson.github.io/research_blog/2022/11/18/TRAPS-debugging-part3.html