These are some notes about the steps that are being done to convert the old program into one setup for the new system.
The new goes streams do not use the gvar format, and the way we read `raw` images from our grb-box from automated sciences is now via exchange of PNG files. So, there is not requirement for r.in.gvar, gvar_inspector, libgvar, etc.
I think DWR might want to maintain the zipcodes, but we will be removing that from the setup completely.
The ca_daily_vis is designed mostly so users can rsync only the visible data. However, since the GOES data is more accessible now, I think we can just use the PNG files as our method of exchanging missed data.
We will still maintain a system where there operation of CIMIS project being run in the /home/cimis directory. This is primarily as a way to be able to update the system as a regular user, esp for the DWR setup, although that is more difficult because of their lack of internet connectivity.
r.sun beam_rad=beam_rad diff_rad=diff_rad refl_rad=refl_rad glob_rad=glob_rad \
day=$(date --date=$(g.gisenv MAPSET) +%j) elevation=Z@500mNow, these should be compared with the r.solpos outputs.
cd .. r.heliosat year=2018 month=06 day=15 elevin=Z@500m linkein=linke3 latitude=latitude ssha=ssh a
One big difference w.r.t r.sun is the fact that r.sun using the horizon to find out when the local sunrise was. Previously, we had decided not to include this, but with 500m data, this is probably something we need to revisit. I’m going out on a limb and suggesting that this might be what takes r.sun awhile to run.
r.horizon elevation=Z output=horangle step=5 bufferzone=200 maxdistance=5000
Mui sent me a mosaiced NED 30m elevation map. I will be using this as the basis for an updated cimis elevation map. I’ll project to EPSG, take an average of the pixels, and then project again to the standard 500m datasets.
# Set Processing mapset g.mapset location=cimis mapset=Z # Project into 3310 with the default region g.region `r.proj location=ll mapset=PERMANENT input=Z_30m method=lanczos output=Z_30m -g` r.proj location=ll mapset=PERMANENT input=Z_30m method=lanczos output=Z_30m # Get Average over 30*15=450m r.neighbors input=Z_30m method=average size=15 \ output=Z_30mx15 title="Average Elevation" # Also, we can do a 500m map here? g.region -d r.mapcalc expression='Z_500m=Z_30mx15' # Change to 500m MAPSET g.mapset 500m # Get our default prodjection g.region -d r.proj location=cimis mapset=Z input=Z_30mx15 output=Z method=lanczos_f
Yeah, comparing Z_500m@Z and Z@500m seem to indicate that they are both pretty much the same, and we could probably get away with using either.
Here are some notes on the EU new Linke Turbidity setup, but reading through these it seems like for CA, the best course of action is to use the MODIS aerosol optical depth measurements. This means, that we’ll need to do some modis downloads, hopefully similar to the sugarcane modis stuff. For now, however, I’m going to use our old data. For now, that means using the same
http://www.soda-pro.com/help/general-knowledge/linke-turbidity-factor