Repo for machine learning models for regional prediction of hydrologic forcing functions. Includes probabilistic seasonal high flow regions for CONUS, and prediction of high flow metrics for selected regions.
Creative Commons Zero v1.0 Universal
0
stars
4
forks
source link
Add NHD attributes that are equivalent to GAGESII attributes #167
We had been training models with some GAGESII dataset attributes. That will not allow us to make predictions for all NHD reaches because those attributes will not be available.
There are similar NHD attributes available for most of these:
Freshwater withdrawal - is this available?
NPDES - we can pull this additional dataset.
development (devl) is already captured from the SOHL data (highly correlated)
Canal and ditch data are already pulled, but seemed better from GAGESII compared to the NHD data (fewer 0s in the record). We can switch back to the NHD dataset for prediction in all reaches.
For now, I'll re-run our models without these GAGESII data. I'm not expecting any differences for our multiclass prediction model because these attributes were not in the top most important.
We had been training models with some GAGESII dataset attributes. That will not allow us to make predictions for all NHD reaches because those attributes will not be available.
There are similar NHD attributes available for most of these:
For now, I'll re-run our models without these GAGESII data. I'm not expecting any differences for our multiclass prediction model because these attributes were not in the top most important.