Usage Guide

Introduction

mapchete-eo extends Mapchete with Earth Observation specific capabilities. This guide covers common patterns for data discovery and processing.

Data Discovery (Search)

Search for data using Source objects.

from mapchete_eo.source import Source

source = Source(
    collection="sentinel-2-l2a",
    catalog_type="stac",
    # ... other config
)

# Search for items
items = list(source.search(area=roi, time="2023-01-01/2023-01-31"))

Reading Data

Products are used to read data into xarray or numpy cubes.

# Inside a mapchete process
def execute(mp, **kwargs):
    with mp.open("sentinel-2") as s2:
        # Read as xarray
        ds = s2.read(eo_bands=["red", "green", "blue"])

        # Read as MaskedArray
        arr = s2.read_np_array(eo_bands=["nir"])

Compositing and Blending

You can use various blending methods to combine overlapping products.

from mapchete_eo.image_operations.compositing import composite

# Composite foreground and background using 'multiply'
blended = composite("multiply", bg_arr, fg_arr)

Example Workflow

Define your tileset and process area in a .mapchete file.
Use mapchete-eo drivers (e.g., sentinel2) to declare inputs.
Access and process Earth Observation data with high-level API functions.