import os
import xarray
import re
import pandas as pd
import json
import tempfile
import boto3
from datetime import datetime
TM5-4DVar Isotopic CH₄ Inverse Fluxes
Documentation of data transformation
This script was used to transform the TM5-4DVar Isotopic CH₄ Inverse Fluxes dataset from netCDF to Cloud Optimized GeoTIFF (COG) format for display in the Greenhouse Gas (GHG) Center.
= boto3.session.Session()
session = session.client("s3")
s3_client = (
bucket_name "ghgc-data-store-dev" # S3 bucket where the COGs are stored after transformation
)= "tm5-ch4-inverse-flux"
FOLDER_NAME
= pd.DataFrame(
files_processed =["file_name", "COGs_created"]
columns# A dataframe to keep track of the files that we have transformed into COGs
)
# Reading the raw netCDF files from local machine
for name in os.listdir(FOLDER_NAME):
= xarray.open_dataset(f"{FOLDER_NAME}/{name}", engine="netcdf4")
xds = xds.rename({"latitude": "lat", "longitude": "lon"})
xds = xds.assign_coords(lon=(((xds.lon + 180) % 360) - 180)).sortby("lon")
xds = [var for var in xds.data_vars if "global" not in var]
variable
for time_increment in range(0, len(xds.months)):
= name.split("/ ")[-1]
filename = re.split("[_ .]", filename)
filename_elements = datetime(int(filename_elements[-2]), time_increment + 1, 1)
start_time for var in variable:
= getattr(xds.isel(months=time_increment), var)
data = data.isel(lat=slice(None, None, -1))
data "lon", "lat", inplace=True)
data.rio.set_spatial_dims("epsg:4326", inplace=True)
data.rio.write_crs(
# # insert date of generated COG into filename
filename_elements.pop()-1] = start_time.strftime("%Y%m")
filename_elements[2, var)
filename_elements.insert(= "_".join(filename_elements)
cog_filename # # add extension
= f"{cog_filename}.tif"
cog_filename
with tempfile.NamedTemporaryFile() as temp_file:
data.rio.to_raster(
temp_file.name,="COG",
driver
)
s3_client.upload_file(=temp_file.name,
Filename=bucket_name,
Bucket=f"{FOLDER_NAME}/{cog_filename}",
Key
)
= files_processed._append(
files_processed "file_name": name, "COGs_created": cog_filename},
{=True,
ignore_index
)
print(f"Generated and saved COG: {cog_filename}")
# Generate the json file with the metadata that is present in the netCDF files.
with tempfile.NamedTemporaryFile(mode="w+") as fp:
json.dump(xds.attrs, fp)"data_dimensions": dict(xds.dims)}, fp)
json.dump({"data_variables": list(xds.data_vars)}, fp)
json.dump({
fp.flush()
s3_client.upload_file(=fp.name,
Filename=bucket_name,
Bucket=f"{FOLDER_NAME}/metadata.json",
Key
)
# creating the csv file with the names of files transformed.
files_processed.to_csv(f"s3://{bucket_name}/{FOLDER_NAME}/files_converted.csv",
)print("Done generating COGs")