The following content is a read-only preview of an executable Jupyter notebook.To run this notebook interactively:
- Go to Wherobots Cloud.
- Start a runtime.
- Open the notebook.
- In the Jupyter Launcher:
- Click File > Open Path.
- Paste the following path to access this notebook:
examples/Analyzing_Data/Zonal_Stats_ESAWorldCover_Texas.ipynb - Click Enter.
- A ESA World Cover raster dataset loaded from AWS Open Data
- Overture Maps Building footprints with a 30 meter buffer in Dallas and the entire state of Texas
from sedona.spark import *
# Create or get a Wherobots session
config = SedonaContext.builder().getOrCreate()
# Initialize the Sedona context which powers spatial processing
sedona = SedonaContext.create(config)
# ✅ Sedona session ready!
# Let's ingest the ESA WorldCover collection. This DataFrame includes metadata and URLs for each raster band.
stac_df = sedona.read.format("stac").load(
"https://services.terrascope.be/stac/collections/urn:eop:VITO:ESA_WorldCover_10m_2021_AWS_V2"
)
stac_df.printSchema()
# ✅ STAC items loaded!
🔑 Key Fields in stac_df:
id: Unique STAC item identifier 🆔properties: Scene metadata (acquisition date, cloud cover, etc.) 📅☁️assets: Map of band names to asset objects (including URLs) 🎞️
# Load the ESA WorldCover data directly from S3
esa = sedona.read.format("raster")\
.option("retile", "true")\
.load("s3://esa-worldcover/v200/2021/map/*.tif*")
esa.show()
# Create a new catalog in Wherobots
esa.createOrReplaceTempView("esa_outdb_rasters")
YOUR_DATABASE_NAME = 'esa'
sedona.sql(f"""CREATE DATABASE IF NOT EXISTS examples_temp.{YOUR_DATABASE_NAME}""")
# Persist as a raster table in Iceberg
sedona.sql(f"""
CREATE OR REPLACE TABLE examples_temp.{YOUR_DATABASE_NAME}.esa_world_cover
AS
SELECT * FROM esa_outdb_rasters
""")
# ✅ Out‑DB raster table created!
Filter your data
Select the bounding box you want to use. The top box is for the entire state of Texas, the second one (uncommented) is for a smaller sample area around Dallas.# tx = 'POLYGON((-106.5715 31.8659,-106.5042 31.7504,-106.3092 31.6242,-106.2103 31.4638,-106.0181 31.3912,-105.7874 31.1846,-105.5663 31.0012,-105.4015 30.8456,-105.0032 30.6462,-104.8521 30.3847,-104.7437 30.2591,-104.6915 30.0738,-104.6777 29.9169,-104.5679 29.7644,-104.5280 29.6475,-104.4044 29.5603,-104.2067 29.4719,-104.1559 29.3834,-103.9774 29.2948,-103.9128 29.2804,-103.8208 29.2481,-103.5640 29.1378,-103.4692 29.0682,-103.3154 29.0105,-103.1616 28.9601,-103.0957 29.0177,-103.0298 29.1330,-102.8677 29.2157,-102.8979 29.2565,-102.8375 29.3570,-102.8004 29.4898,-102.7002 29.6881,-102.5134 29.7691,-102.3843 29.7596,-102.3047 29.8788,-102.1509 29.7834,-101.7004 29.7572,-101.4917 29.7644,-101.2939 29.6308,-101.2582 29.5269,-101.0056 29.3642,-100.9204 29.3056,-100.7707 29.1642,-100.7007 29.0946,-100.6306 28.9012,-100.4974 28.6593,-100.3601 28.4675,-100.2969 28.2778,-100.1733 28.1882,-100.0195 28.0526,-99.9344 27.9435,-99.8438 27.7638,-99.7119 27.6641,-99.4812 27.4839,-99.5375 27.3059,-99.4290 27.1948,-99.4455 27.0175,-99.3164 26.8829,-99.2065 26.6867,-99.0967 26.4116,-98.8138 26.3574,-98.6668 26.2257,-98.5474 26.2343,-98.3276 26.1357,-98.1697 26.0457,-97.9143 26.0518,-97.6643 26.0050,-97.4020 25.8419,-97.3526 25.9074,-97.0148 25.9679,-97.0697 26.1789,-97.2249 26.8253,-97.0752 27.4230,-96.6096 28.0599,-95.9285 28.4228,-95.3036 28.7568,-94.7296 29.0742,-94.3355 29.3810,-93.8205 29.6021,-93.9317 29.8013,-93.8136 29.9157,-93.7230 30.0489,-93.6996 30.1214,-93.7216 30.2021,-93.7038 30.2792,-93.7628 30.3278,-93.7587 30.3835,-93.7010 30.4380,-93.7024 30.5079,-93.7299 30.5362,-93.6694 30.6296,-93.6090 30.7466,-93.5527 30.8114,-93.5747 30.8834,-93.5307 30.9376,-93.5074 31.0318,-93.5266 31.0812,-93.5335 31.1787,-93.5980 31.1670,-93.6832 31.3055,-93.6708 31.3830,-93.6887 31.4369,-93.7202 31.5107,-93.8315 31.5820,-93.8123 31.6440,-93.8232 31.7188,-93.8342 31.7936,-93.8782 31.8309,-93.9221 31.8869,-93.9661 31.9335,-94.0430 32.0081,-94.0430 33.4681,-94.0430 33.5414,-94.1528 33.5689,-94.1968 33.5872,-94.2627 33.5872,-94.3176 33.5689,-94.3945 33.5597,-94.4275 33.5780,-94.4275 33.6055,-94.4495 33.6421,-94.4879 33.6329,-94.5236 33.6421,-94.6637 33.6695,-94.7461 33.7061,-94.8999 33.7791,-95.0757 33.8818,-95.1526 33.9251,-95.2254 33.9604,-95.2858 33.8750,-95.5399 33.8841,-95.7568 33.8887,-95.8420 33.8408,-96.0274 33.8556,-96.3528 33.6901,-96.6179 33.8442,-96.5836 33.8898,-96.6673 33.8955,-96.7538 33.8179,-96.8335 33.8613,-96.8774 33.8613,-96.9159 33.9388,-97.0917 33.7392,-97.1645 33.7449,-97.2180 33.8978,-97.3746 33.8225,-97.4611 33.8305,-97.4460 33.8761,-97.6945 33.9798,-97.8648 33.8476,-97.9651 33.8978,-98.0983 34.0299,-98.1752 34.1141,-98.3743 34.1425,-98.4773 34.0640,-98.5529 34.1209,-98.7520 34.1232,-98.9539 34.2095,-99.0637 34.2073,-99.1832 34.2141,-99.2505 34.3593,-99.3823 34.4613,-99.4318 34.3774,-99.5718 34.4160,-99.6158 34.3706,-99.8094 34.4726,-99.9934 34.5631,-100.0017 36.4975,-103.0408 36.5008,-103.0655 32.0011,-106.6168 32.0023,-106.5715 31.8659))'
# Dallas Only
tx = 'POLYGON((-97.000482 33.023937, -96.463632 33.023937, -96.463632 32.613222, -97.000482 32.613222, -97.000482 33.023937))'
buildings = sedona.read.format("iceberg").load(
"wherobots_open_data.overture_maps_foundation.buildings_building")
buildings.createOrReplaceTempView('buildings')
buildings_filtered = sedona.sql(f'''select *,
st_buffer(st_transform(geometry, 'epsg:4326', 'epsg:3857'), 30) as buffer
from buildings
where st_intersects(geometry, st_geomfromtext('{tx}'))''')
buildings_filtered = buildings_filtered.repartition(200)
buildings_filtered.createOrReplaceTempView('buildings_filtered')
buildings_filtered.count()
# ✅ Isochrone table loaded!
# Persist buildings as a raster table in Iceberg
sedona.sql(f"""
CREATE OR REPLACE TABLE examples_temp.{YOUR_DATABASE_NAME}.texas_buildings
AS
SELECT * FROM buildings_filtered
""")
Run zonal stats
zonal_stats_df = sedona.sql(f"""
SELECT /*+ BROADCAST(p) */
p.id,
RS_ZonalStatsAll(r.rast, p.buffer, 1) AS stats
FROM
examples_temp.{YOUR_DATABASE_NAME}.texas_buildings p
JOIN
examples_temp.{YOUR_DATABASE_NAME}.esa_world_cover r
ON RS_Intersects(r.rast, p.buffer)
""")
# ✅ Zonal stats computed!
import os
user_uri = os.getenv("USER_S3_PATH")
%%timeit
# Creates a loop to test the performance using a fake file writer to showthe output of the entire dataset
zonal_stats_df.write.format("noop").mode("overwrite").save(user_uri + "/fakeOutput")