{
"cells": [
{
"cell_type": "markdown",
"id": "ee0441c8",
"metadata": {},
"source": [
"# Getting Started\n",
"The Chalmers Atmospheric Water Dataset from the Arctic Weather Satellite (AWS), provides global atmospheric ice mass estimates derived from passive microwave measurements.\n",
"The data are available as NetCDF files, the following examples will show how to get started using them with python and `xarray`."
]
},
{
"cell_type": "code",
"execution_count": null,
"id": "90be1304",
"metadata": {},
"outputs": [],
"source": [
"from datetime import datetime\n",
"import requests\n",
"import xarray as xr\n",
"\n",
"# Helper function to get file urls for given timeranges\n",
"def file_urls_for_timeranges(timeranges: list[slice]) -> list[str]:\n",
" \"\"\"\n",
" Get a list of level2 urls for the requested timerange\n",
" \"\"\"\n",
" base_url = \"https://storage.googleapis.com/petermfiles/cawd-aws-v0.0.1-example/level2\"\n",
" index = requests.get(f\"{base_url}/index.json\").json()\n",
" urls = []\n",
" for timerange in timeranges:\n",
" tr_slice = slice(\n",
" datetime.fromisoformat(timerange.start.replace(\"Z\", \"+00:00\")),\n",
" datetime.fromisoformat(timerange.stop.replace(\"Z\", \"+00:00\"))\n",
" )\n",
" urls.extend([\n",
" f\"{base_url}/{item['filepath']}#mode=bytes\"\n",
" for item in index['items']\n",
" if (datetime.fromisoformat(item['datetime_end']) >= tr_slice.start)\n",
" and (datetime.fromisoformat(item['datetime_start']) <= tr_slice.stop)\n",
" ])\n",
" return sorted(set(urls))"
]
},
{
"cell_type": "code",
"execution_count": 2,
"id": "515150c0",
"metadata": {},
"outputs": [
{
"data": {
"text/html": [
"
Input Top of Atmosphere Brightness Temperature values used for retrieval.
\n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Array
\n",
"
Chunk
\n",
"
\n",
" \n",
" \n",
" \n",
"
\n",
"
Bytes
\n",
"
24.84 MiB
\n",
"
12.42 MiB
\n",
"
\n",
" \n",
"
\n",
"
Shape
\n",
"
(6727, 88, 11)
\n",
"
(3364, 88, 11)
\n",
"
\n",
"
\n",
"
Dask graph
\n",
"
2 chunks in 5 graph layers
\n",
"
\n",
"
\n",
"
Data type
\n",
"
float32 numpy.ndarray
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
sim_db_all_fwp
(time, fov)
float32
dask.array<chunksize=(3364, 88), meta=np.ndarray>
units :
1
long_name :
Fwp values of the closest observation in the simulation database in terms of channel temperatures.
\n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Array
\n",
"
Chunk
\n",
"
\n",
" \n",
" \n",
" \n",
"
\n",
"
Bytes
\n",
"
2.26 MiB
\n",
"
1.13 MiB
\n",
"
\n",
" \n",
"
\n",
"
Shape
\n",
"
(6727, 88)
\n",
"
(3364, 88)
\n",
"
\n",
"
\n",
"
Dask graph
\n",
"
2 chunks in 5 graph layers
\n",
"
\n",
"
\n",
"
Data type
\n",
"
float32 numpy.ndarray
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
sim_db_all_fwp_dm
(time, fov)
float32
dask.array<chunksize=(3364, 88), meta=np.ndarray>
units :
1
long_name :
Fwp_Dm values of the closest observation in the simulation database in terms of channel temperatures.
\n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Array
\n",
"
Chunk
\n",
"
\n",
" \n",
" \n",
" \n",
"
\n",
"
Bytes
\n",
"
2.26 MiB
\n",
"
1.13 MiB
\n",
"
\n",
" \n",
"
\n",
"
Shape
\n",
"
(6727, 88)
\n",
"
(3364, 88)
\n",
"
\n",
"
\n",
"
Dask graph
\n",
"
2 chunks in 5 graph layers
\n",
"
\n",
"
\n",
"
Data type
\n",
"
float32 numpy.ndarray
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
sim_db_all_fwp_zm
(time, fov)
float32
dask.array<chunksize=(3364, 88), meta=np.ndarray>
units :
1
long_name :
Fwp_Zm values of the closest observation in the simulation database in terms of channel temperatures.
\n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Array
\n",
"
Chunk
\n",
"
\n",
" \n",
" \n",
" \n",
"
\n",
"
Bytes
\n",
"
2.26 MiB
\n",
"
1.13 MiB
\n",
"
\n",
" \n",
"
\n",
"
Shape
\n",
"
(6727, 88)
\n",
"
(3364, 88)
\n",
"
\n",
"
\n",
"
Dask graph
\n",
"
2 chunks in 5 graph layers
\n",
"
\n",
"
\n",
"
Data type
\n",
"
float32 numpy.ndarray
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
sim_db_all_ta_distance
(time, fov)
float32
dask.array<chunksize=(3364, 88), meta=np.ndarray>
units :
K
long_name :
"Distance" to the closest observation in the simulation database in terms of channel temperatures.
\n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Array
\n",
"
Chunk
\n",
"
\n",
" \n",
" \n",
" \n",
"
\n",
"
Bytes
\n",
"
2.26 MiB
\n",
"
1.13 MiB
\n",
"
\n",
" \n",
"
\n",
"
Shape
\n",
"
(6727, 88)
\n",
"
(3364, 88)
\n",
"
\n",
"
\n",
"
Dask graph
\n",
"
2 chunks in 5 graph layers
\n",
"
\n",
"
\n",
"
Data type
\n",
"
float32 numpy.ndarray
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
l1b_index_scans
(time)
uint32
dask.array<chunksize=(3364,), meta=np.ndarray>
units :
1
long_name :
Index along the n_scans dimension in orignal L1B file (along-track). In the L2 file, scan time is sued as the along-track dimension in order to facilitate easier concatination and selection based on timeranges.
\n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Array
\n",
"
Chunk
\n",
"
\n",
" \n",
" \n",
" \n",
"
\n",
"
Bytes
\n",
"
26.28 kiB
\n",
"
13.14 kiB
\n",
"
\n",
" \n",
"
\n",
"
Shape
\n",
"
(6727,)
\n",
"
(3364,)
\n",
"
\n",
"
\n",
"
Dask graph
\n",
"
2 chunks in 5 graph layers
\n",
"
\n",
"
\n",
"
Data type
\n",
"
uint32 numpy.ndarray
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
l1b_index_fovs
(time, fov)
uint32
dask.array<chunksize=(3364, 88), meta=np.ndarray>
units :
1
long_name :
Index along n_fovs dimension in orignal L1B file
\n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Array
\n",
"
Chunk
\n",
"
\n",
" \n",
" \n",
" \n",
"
\n",
"
Bytes
\n",
"
2.26 MiB
\n",
"
1.13 MiB
\n",
"
\n",
" \n",
"
\n",
"
Shape
\n",
"
(6727, 88)
\n",
"
(3364, 88)
\n",
"
\n",
"
\n",
"
Dask graph
\n",
"
2 chunks in 7 graph layers
\n",
"
\n",
"
\n",
"
Data type
\n",
"
uint32 numpy.ndarray
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
flag_bad_data
(time, fov)
uint8
dask.array<chunksize=(3364, 88), meta=np.ndarray>
units :
1
flag_values :
0, other
flag_meanings :
bit 0: bad_ta_value, bit 1: bad_satellite_altitude, bit 2: bad_latlon, bit 3: bad_aa_za, bit 4: bad_navigation_status, bit 5: bad_scanline_quality, bit 6: bad_aws_l1b_processing_info_chan
long_name :
Quality flag for observation (scan and fov). 0 if values in scan pass basic QA checks. Other if any of the checks fail. Bit position indicates which check failed.
\n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Array
\n",
"
Chunk
\n",
"
\n",
" \n",
" \n",
" \n",
"
\n",
"
Bytes
\n",
"
578.10 kiB
\n",
"
289.09 kiB
\n",
"
\n",
" \n",
"
\n",
"
Shape
\n",
"
(6727, 88)
\n",
"
(3364, 88)
\n",
"
\n",
"
\n",
"
Dask graph
\n",
"
2 chunks in 5 graph layers
\n",
"
\n",
"
\n",
"
Data type
\n",
"
uint8 numpy.ndarray
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
flag_overlap
(time)
bool
dask.array<chunksize=(3364,), meta=np.ndarray>
units :
1
flag_values :
0, 1
flag_meanings :
Scan is not present in previous L1b file, Scan is present in previous L1b file
long_name :
Flag for indicating that scan is present in previous L1B file
\n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
Array
\n",
"
Chunk
\n",
"
\n",
" \n",
" \n",
" \n",
"
\n",
"
Bytes
\n",
"
6.57 kiB
\n",
"
3.29 kiB
\n",
"
\n",
" \n",
"
\n",
"
Shape
\n",
"
(6727,)
\n",
"
(3364,)
\n",
"
\n",
"
\n",
"
Dask graph
\n",
"
2 chunks in 5 graph layers
\n",
"
\n",
"
\n",
"
Data type
\n",
"
bool numpy.ndarray
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
\n",
" \n",
"
\n",
"
\n",
"
fwp_ccic
(time, fov)
float32
dask.array<chunksize=(3364, 88), meta=np.ndarray>
units :
kg/m2
long_name :
Co-located FWP from the Chalmers Cloud Ice Climatology (ccic.readthedocs.io) for comparison.