search

r4atlantis / atlantisom

Atlantis operating model. Generates data sets from Atlantis scenarios.
https://r4atlantis.github.io/atlantisom
8 stars 6 forks source link

function load_atlantis_ncdf #3

Closed kellijohnson-NOAA closed 8 years ago

kellijohnson-NOAA commented 8 years ago

load_atlantis_ncdf is currently written but has some issues:

kellijohnson-NOAA commented 8 years ago

@alketh can you provide access to get_bboxes()

alketh commented 8 years ago

get_bboxes() is hard coded in the current state of my package. As far as I know, currently there is no way to extract the boundary boxes in an automated way. We have to check with Beth to solve this.

kellijohnson-NOAA commented 8 years ago

the hard coded portion in your package: is it just a vector? or what are you passing inside load_atlantis_nncdf?

alketh commented 8 years ago

The output of get_bboxes() is simply a vector of integers given the id of the boundary boxes. For the CaCu model these are: 0, 54, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88

slucey commented 8 years ago

We can make it a simple argument for the time being.

Sean

Sean M. Lucey Fisheries Biologist U. S. Dept. of Commerce/NOAA/NMFS Northeast Fisheries Science Center 166 Water Street, Woods Hole, MA 02543 508-495-2011 (voice) 508-495-2232 (fax)

On Mon, Dec 7, 2015 at 2:47 PM, Alexander Keth notifications@github.com wrote:

The output of get_bboxes() is simply a vector of integers given the id of the boundary boxes. For the CaCu model these are: 0, 54, 75, 76, 77, 78, 79, 80, 81, 82, 83, 84, 85, 86, 87, 88

— Reply to this email directly or view it on GitHub https://github.com/r4atlantis/atlantisom/issues/3#issuecomment-162638486 .

kellijohnson-NOAA commented 8 years ago

i almost have a function, get_boundary to get it from the .bgm file, it is currently picking up every boundary box except 1, box 54, which is an island.

alketh commented 8 years ago

Treating islands as boundary boxes should work fine.

kellijohnson-NOAA commented 8 years ago

I cannot pick up the island with my function though. I will have to ask Beth, Bec, or Isaac on how to find islands, maybe something regarding if land is included and if it is not assume all boxes on land are boundary?

On 12/7/2015 12:01 PM, Alexander Keth wrote:

Treating islands as boundary boxes should work fine.

— Reply to this email directly or view it on GitHub https://github.com/r4atlantis/atlantisom/issues/3#issuecomment-162643632.

alketh commented 8 years ago

Sorry guys, I am having a hard time to get used to RNetCDF due to the complete different structure of the loaded nc-files. How can I extract the number if dimensions of a variable in the ncfile?

kellijohnson-NOAA commented 8 years ago

RNetCDF::dim.inq.nc(at_out, 0)$length will give you the dimensions of timesteps. In the command line run: git fetch git rebase origin/master to get my latest changes, as I already fixed a few things.

On 12/7/2015 12:04 PM, Alexander Keth wrote:

Sorry guys, I am having a hard time to getting used to RNetCDF due to the complete different structure of the loaded nc-files. How can I extract the number if dimensions of a variable in the ncfile?

— Reply to this email directly or view it on GitHub https://github.com/r4atlantis/atlantisom/issues/3#issuecomment-162644361.

kellijohnson-NOAA commented 8 years ago

@alketh what file are you referring to when you get the biomass pools? The CaCur model does not have an init...nc file.

alketh commented 8 years ago

The file is named "DIVCalCurrentV3_Biol.nc". Isaac didn't supply it with his model-folder. I make sure that everybody in here gets a copy of it.

kellijohnson-NOAA commented 8 years ago

great

On 12/7/2015 04:16 PM, Alexander Keth wrote:

The file is named "DIVCalCurrentV3_Biol.nc". Isaac didn't supply it with his model-folder. I make sure that everybody in here gets a copy of it.

— Reply to this email directly or view it on GitHub https://github.com/r4atlantis/atlantisom/issues/3#issuecomment-162712396.

alketh commented 8 years ago

I just checked the new version of the load_nc function. The code looks great. However, I would like to give a catious remark regarding the file structure of different atlantis models. We have to be cautious when we supply a "global" direction ("dir") of the input files. This does only work if all files are read in from one location. Unfortunately, this is not the case for multiple atlantis models. Many models have a main folder and output folder. Therefore I highly suggest to provide the files as full folder/filename string (similar to the "con" parameter used in various R-functions).

ghost commented 8 years ago

Hey all,

I am working on the calc_stage2age and in order to do that I needed an example file -- when I run the load_nc file the column of time just outputs zeros no matter what run's nc file I am pulling in. Have others found this?

alketh commented 8 years ago

Hey,

I just read in "ResN" from the general output-file using load_nc. All time steps (from 0 to 63) were present in the resulting dataframe.

Am 08.12.2015 um 15:05 schrieb hodgsone:

Hey all,

I am working on the calc_stage2age and in order to do that I needed an example file -- when I run the load_nc file the column of time just outputs zeros no matter what run's nc file I am pulling in. Have others found this?

— Reply to this email directly or view it on GitHub https://github.com/r4atlantis/atlantisom/issues/3#issuecomment-163074286.

<°)))>< ><(((°> <°)))>< ><(((°> <°)))>< ><(((°>

Alexander Keth Institute for Hydrobiology and Fisheries Science University of Hamburg Olbersweg 24 22767 Hamburg Germany Tel.: +49 (0)40 428 38 6708 Fax: +49 (0)40 428 38 6618

<°)))>< ><(((°> <°)))>< ><(((°> <°)))>< ><(((°>

slucey commented 8 years ago

I have some sample data we can use but GitHub won't let me send it through this e-mail string. Send me your e-mail address and I'll send it to you.

Sean

Sean M. Lucey Fisheries Biologist U. S. Dept. of Commerce/NOAA/NMFS Northeast Fisheries Science Center 166 Water Street, Woods Hole, MA 02543 508-495-2011 (voice) 508-495-2232 (fax)

On Tue, Dec 8, 2015 at 8:13 PM, Alexander Keth notifications@github.com wrote:

Hey,

I just read in "ResN" from the general output-file using load_nc. All time steps (from 0 to 63) were present in the resulting dataframe.

Am 08.12.2015 um 15:05 schrieb hodgsone:

Hey all,

I am working on the calc_stage2age and in order to do that I needed an example file -- when I run the load_nc file the column of time just outputs zeros no matter what run's nc file I am pulling in. Have others found this?

— Reply to this email directly or view it on GitHub < https://github.com/r4atlantis/atlantisom/issues/3#issuecomment-163074286>.

<°)))>< ><(((°> <°)))>< ><(((°> <°)))>< ><(((°>

Alexander Keth Institute for Hydrobiology and Fisheries Science University of Hamburg Olbersweg 24 22767 Hamburg Germany Tel.: +49 (0)40 428 38 6708 Fax: +49 (0)40 428 38 6618

<°)))>< ><(((°> <°)))>< ><(((°> <°)))>< ><(((°>

— Reply to this email directly or view it on GitHub https://github.com/r4atlantis/atlantisom/issues/3#issuecomment-163075287 .

ghost commented 8 years ago

Thanks Sean -- my email is hodgsone@uw.edu, but I also have sample files so I am not sure that is the issue -- these are outputs from old runs I have done and have multiple time steps. I am not sure why it does not work for mine. I can try again

kellijohnson-NOAA commented 8 years ago

load_nc works great.