test_sentinelsat

Using Python to search and download Sentinel data¶

In [1]:

from sentinelsat import SentinelAPI, read_geojson, geojson_to_wkt
import xarray
import pandas as pd

“Sentinelsat makes searching, downloading and retrieving the metadata of Sentinel satellite images from the Copernicus Open Access Hub easy.” From Sentinelsat — Sentinelsat 1.1.1 documentation

connect to the Python API

In [2]:

api = SentinelAPI('username', 'password', 'https://apihub.copernicus.eu/apihub')

search by polygon, time, and SciHub query keywords

I’ve created a simple ROI (rectangle) as geojson file, called map.geojson in order to intersect the footprint with data to search.

In [3]:

footprint = geojson_to_wkt(read_geojson('../sentinel1/map.geojson'))
products = api.query(footprint,
                     date = ('20180605', '20180610'),
                     platformname = 'Sentinel-1',
                     producttype='SLC')

A SentinelAPI object can allow different parameters. Here it is date of ingestion , the type of platform and wihich kkind of product you need.

convert to Pandas DataFrame

In [5]:

products_df = api.to_dataframe(products)

In [ ]:

products_df

Testing products_df visualization as xarray¶

Testing Download API by using the id product¶

Before download sentinel data, control if the product is Online and able to directly download. Otherwise you must unblock the product by swutching it from OFFLINE status to ONLINE status.

In [14]:

product_info = api.get_product_odata('8a464f41-cb80-4350-848f-a95da9810991')
product_info['Online']

Out[14]:

False

In [39]:

api.download('8a464f41-cb80-4350-848f-a95da9810991')

Downloading S1B_IW_SLC__1SDV_20180607T041420_20180607T041448_011262_014ABB_AC0E.zip:   0%|          | 0.00/4.5…

MD5 checksumming:   0%|          | 0.00/4.50G [00:00<?, ?B/s]

Out[39]:

{'id': '8a464f41-cb80-4350-848f-a95da9810991',
 'title': 'S1B_IW_SLC__1SDV_20180607T041420_20180607T041448_011262_014ABB_AC0E',
 'size': 4496221397,
 'md5': 'cd13a4fe35c5c197ca83c75c7acf036f',
 'date': datetime.datetime(2018, 6, 7, 4, 14, 20, 208000),
 'footprint': 'POLYGON((28.414934 37.326553,25.574972 37.724804,25.919802 39.401939,28.828909 39.004982,28.414934 37.326553))',
 'url': "https://apihub.copernicus.eu/apihub/odata/v1/Products('8a464f41-cb80-4350-848f-a95da9810991')/$value",
 'Online': True,
 'Creation Date': datetime.datetime(2018, 6, 7, 9, 21, 30, 469000),
 'Ingestion Date': datetime.datetime(2018, 6, 7, 9, 13, 12, 2000),
 'quicklook_url': "https://apihub.copernicus.eu/apihub/odata/v1/Products('8a464f41-cb80-4350-848f-a95da9810991')/Products('Quicklook')/$value",
 'path': 'S1B_IW_SLC__1SDV_20180607T041420_20180607T041448_011262_014ABB_AC0E.zip',
 'downloaded_bytes': 4496221397}

In this case I select a product offline. It exists an API (trigger_offline_retrieval) in which we pass the id product in order to re-activate the product in 24h. The response of server is a boolean (True) for the success of request. Then you could try to download your desired data.

In [6]:

api.trigger_offline_retrieval('8a464f41-cb80-4350-848f-a95da9810991')

Out[6]:

True

Displaying product info¶

In [16]:

products_df.info()

<class 'pandas.core.frame.DataFrame'>
Index: 4 entries, 87438c76-c585-43d6-875d-490056fab559 to be152722-3387-4d7c-a9fa-0ce74f824e9c
Data columns (total 34 columns):
 #   Column                   Non-Null Count  Dtype         
---  ------                   --------------  -----         
 0   title                    4 non-null      object        
 1   link                     4 non-null      object        
 2   link_alternative         4 non-null      object        
 3   link_icon                4 non-null      object        
 4   summary                  4 non-null      object        
 5   ondemand                 4 non-null      object        
 6   ingestiondate            4 non-null      datetime64[ns]
 7   beginposition            4 non-null      datetime64[ns]
 8   endposition              4 non-null      datetime64[ns]
 9   missiondatatakeid        4 non-null      int64         
 10  orbitnumber              4 non-null      int64         
 11  lastorbitnumber          4 non-null      int64         
 12  relativeorbitnumber      4 non-null      int64         
 13  lastrelativeorbitnumber  4 non-null      int64         
 14  slicenumber              4 non-null      int64         
 15  acquisitiontype          4 non-null      object        
 16  filename                 4 non-null      object        
 17  gmlfootprint             4 non-null      object        
 18  format                   4 non-null      object        
 19  identifier               4 non-null      object        
 20  instrumentshortname      4 non-null      object        
 21  sensoroperationalmode    4 non-null      object        
 22  instrumentname           4 non-null      object        
 23  swathidentifier          4 non-null      object        
 24  footprint                4 non-null      object        
 25  platformidentifier       4 non-null      object        
 26  orbitdirection           4 non-null      object        
 27  polarisationmode         4 non-null      object        
 28  productclass             4 non-null      object        
 29  producttype              4 non-null      object        
 30  platformname             4 non-null      object        
 31  size                     4 non-null      object        
 32  status                   4 non-null      object        
 33  uuid                     4 non-null      object        
dtypes: datetime64[ns](3), int64(6), object(25)
memory usage: 1.1+ KB

If we want to display all the information within the columns of Panda Dataframe, we must set the max number of columns we need to visualize. I set an hight number in order to all visualize.

In [8]:

pd.set_option('display.max_columns', 100)

In [ ]:

products_sorted

Sentinelsat allows filtering and sorting of search results before download.

In [ ]:

products_sorted = products_df.sort_values(['size'], ascending=[True])
products_sorted