- Git
- Docker
- Some python knowledge and jupyter notebooks
Now clone the repository and follow the instructions.
git clone git@github.com:davivc/labar-coral-sol.git
cd labar-coral-sol
In this project we define 3 services for different tasks inside the docker-compose.yml file. Each task will run in its specific container.
- getdata: this service is responsible to download the netCDF files from https://resources.marine.copernicus.eu/.
- regridnetcdf: this service is responsible to regrid the downloaded netCDF files to a smaller scale
- tubastraea_notebook: this service will hold the code and the jupyter notebooks for our model
Warning: To run the codes inside the notebook you must have the regrided netcdf files. So you must run the first 2 containers at least one time to fetch the data and regrid it. After that you can execute the notebook whenever you want.
To run the code we first need download images and build containers. This step will take some time (maybe a couple hours depending on your internet and RAM Memory)
This command will build all 3 containers
docker-compose build
You can also build each one at a time
docker-compose build getdata
docker-compose build regridnetcdf
docker-compose build tabastraea_notebook
To retrieve the data first copy the file .env-example to .env and change the credentials for Copernicus. If you don't have and account you can create here at https://resources.marine.copernicus.eu/?option=com_sla. After that change the credentials in the .env file to your credentials
MOTU_USER=YOUR_USER
MOTU_PWD=YOUR_PASSWORD
With the credentials configured just execute and wait the script to finish downloading. It will take approximately 1 hour (depending your on connection).
docker-compose run --rm getdata
This command will retrieve nc data from 1993 to 2018 for the variables thetao, uo, vo and zos and depth between 0.493 and 21.599. However you can change the file docker/getdata.sh to retrieve different data. After the changes build the container again and execute the command above.
Run regrid to downscale NetCDF files. This step will regrid the nc files from ~9km to 3km, then 3km to 1km and finally will regrid 1km to 500m. It will take approximately 10 minutes on a pc with 16GB RAM.
docker-compose run --rm regridnetcdf regrid.py
Or you can select one year only
docker-compose run --rm regridnetcdf regrid.py YEAR
docker-compose up tubastraea_notebook
After that go to the url shown in your terminal and navigate through the notebooks.
To validate our regrid we will use data from PNBOIA at https://www.marinha.mil.br/chm/dados-do-pnboiaboias/boia-itajai
DMS: Lat 27°24,35'S / Lon 047°15,93'W DD: Lat 27.4° / Lon 47.2655°
To see the parcels advection on action just fire up our stack
[1] Engler R, Guisan A. MigClim: Predicting plant distribution and dispersal in a changing climate. Divers Distrib 2009;15:590–601. doi:10.1111/j.1472-4642.2009.00566.x.
-26.767222 -46.767264
Incorporar desvio padrão da validação Esquecer função da temperatura Assentar protegido cor vermelha Assentar exposto cor amarela Validação Preditivo
Taxa de mortalidade progressiva