Without a Grid File

If a grid file is not available, the system will use tile metadata to perform spatial filtering.

In this approach, you simply replace the specific grid indices (h and v) in the source string with a wildcard (*).

This example demonstrates how to download data for Albania during January 2017 without using a grid file.

Warning

This method can be slower because it first fetches metadata for all available files before applying the bounding box filter.

import earth_data_kit as edk
import datetime

# Define the source path for MODIS scene files using a wildcard pattern.
# The pattern uses strftime formatting (%Y%j) to dynamically match dates within the S3 directory structure.
source = "s3://modis-pds/MCD43A4.006/*/*/%Y%j/*_B07.TIF"

# Instantiate the Dataset object for the MODIS collection using the S3 engine.
# The Dataset class handles scanning and filtering of scene files.
ds = edk.stitching.Dataset("modis-pds", source, "s3")


# Define the spatial extent (in EPSG:4326) using Albania's bounding box.
# Format: (min_longitude, min_latitude, max_longitude, max_latitude)
bbox = (19.3044861183, 39.624997667, 21.0200403175, 42.6882473822)

# Define the temporal range for data selection.
# The dataset covers the month of January 2017; note that the end date is inclusive.
date_range = (
    datetime.datetime(2017, 1, 1),
    datetime.datetime(2017, 1, 31),
)

# Configure the Dataset with the defined time bounds.
ds.set_timebounds(date_range[0], date_range[1])

# Configure the spatial filter for the Dataset.
# Since no grid file is provided, the filtering is based solely on the bounding box.
ds.set_spacebounds(bbox)

# Discover the available scene files that match the defined criteria.
# This step scans the remote source and identifies all files that meet the spatial and temporal constraints.
ds.discover()

# Generate VRTs files for the discovered scene files.
ds.to_vrts(bands=["Nadir_Reflectance_Band7"])