Accessing ICESat-2 Data

Accessing ICESat-2 Data#

Learning Objectives

Use icepyx to search, download, and read ICESat-2 granules
Use sliderule to get GeoDataFrames of ICESat-2 data
Use h5coro to directly read ICESat-2 granules in an S3 bucket

Part 2: SlideRule#

SlideRule is a collaborative effort between NASA Goddard Space Flight Center (GSFC) and the University of Washington, funded by the ICESat-2 program. It provides on-demand science data processing service for ICESat-2 and GEDI data that runs on Amazon Web Services (AWS) and responds to REST-like API calls to process and return science results. This science-data-as-a-service model is a new way for researchers to work and analyze data, enabling them to have low-latency access to custom-generated, high-level data products.

SlideRule users provide specific parameters at the time of the request to compute products that fit their science needs. SlideRule then uses cloud-optimized versions of computational algorithms and a scalable cluster of EC2 instances to process data efficiently. All data is then returned to the user as a geopandas GeoDataFrame.

SlideRule Overview

For more information#

Website: https://slideruleearth.io
Documentation: https://slideruleearth.io/web/rtd/
GitHub: SlideRuleEarth/sliderule
Examples: SlideRuleEarth/sliderule-python
Contact: support@mail.slideruleearth.io

# To use the latest version of the sliderule client, run this cell.
# It will install the sliderule Python client into your current conda environment.
# You will then need to restart your kernel to have the changes take effect.
%pip install --quiet "sliderule>=4.6"

Note: you may need to restart the kernel to use updated packages.

Example 1: Just Get Me Some Data#

# (1) Import the client
from sliderule import sliderule, icesat2

# (2) Initialize the client
sliderule.init("slideruleearth.io");

# (3) Define an area of interest
region = sliderule.toregion("grandmesa.geojson");

# (4) Specify the processing parameters
parms = {
    "poly": region["poly"],
    "srt": icesat2.SRT_LAND,
    "len": 20.0,
    "res": 100.0
}

# (5) Make the processing request
gdf = icesat2.atl06p(parms)

Display the results#

gdf

	region	h_sigma	rms_misfit	spot	pflags	rgt	y_atc	w_surface_window_final	gt	x_atc	h_mean	segment_id	dh_fit_dx	cycle	n_fit_photons	geometry
time
2018-10-16 10:49:21.763047168	6	0.059213	0.414242	3	0	272	41194.648438	3.146032e+00	40	15710061.0	1797.827692	784344	0.121020	1	49	POINT (-108.09813 39.15732)
2018-10-16 10:49:21.773934080	6	0.171685	0.679926	6	0	272	44567.667969	3.917953e+00	10	15712286.0	2205.110892	784455	0.151762	1	17	POINT (-108.06191 39.13431)
2018-10-16 10:49:21.893560064	6	0.530147	1.702927	6	0	272	44545.421875	1.265432e+01	10	15713088.0	2260.114661	784495	0.182269	1	11	POINT (-108.0631 39.12714)
2018-10-16 10:49:22.009937920	6	0.035914	0.287156	1	0	272	37942.792969	1.717312e+01	60	15711985.0	1813.438733	784440	0.059123	1	64	POINT (-108.13779 39.14302)
2018-10-16 10:49:22.116894976	6	0.000000	0.000000	6	1	272	44486.722656	3.000000e+01	10	15714591.0	2549.139896	784570	-0.791383	1	23	POINT (-108.06554 39.11371)
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
2024-05-05 21:42:08.529739776	2	6.008234	308.324646	1	0	737	-1859.211670	3.588987e+20	10	4352867.5	1948.284281	217054	2.667867	23	2689	POINT (-108.09843 39.12156)
2024-05-05 21:42:08.594035200	2	5.008744	204.956665	4	0	737	1433.140137	3.588987e+20	40	4355633.5	1780.562661	217192	-0.860279	23	1675	POINT (-108.13939 39.14353)
2024-05-05 21:42:08.608098560	2	0.000000	0.000000	4	1	737	1433.046631	1.914062e+01	40	4355734.0	1777.159349	217197	-0.750229	23	1698	POINT (-108.1395 39.14443)
2024-05-05 21:42:08.664648704	2	4.936912	201.764343	4	0	737	1432.968018	3.588987e+20	40	4356135.0	2167.969007	217217	-1.484943	23	1671	POINT (-108.13994 39.14802)
2024-05-05 21:42:08.881896704	2	4.585921	200.818405	1	0	737	-1858.465698	3.588987e+20	10	4355373.0	1866.818607	217179	1.150830	23	1918	POINT (-108.1012 39.14401)

15904 rows × 16 columns

Plot the results#

import matplotlib.pyplot as plt

region_lon = [e["lon"] for e in region["poly"]]
region_lat = [e["lat"] for e in region["poly"]]

f, ax = plt.subplots()
ax.set_title("ATL06-SR Points")
ax.set_aspect('equal')
gdf.plot(ax=ax, column='h_mean', cmap='inferno', s=0.1)
ax.plot(region_lon, region_lat, linewidth=1, color='g');

plt.show()

../../_images/a538d610cda7b82eb91d762ad7639a587aae106478e5d60716c3cba1564503a3.png

Explanation of what happened#

(1) Import the client#

from sliderule import sliderule, icesat2

The SlideRule Python client is broken up into different modules:

sliderule: core general functionality
icesat2: ICESat-2 on-demand, subsetting, raster sampling products
gedi: GEDI subsetting, and raster sampling products
h5: direct HDF5 data access
earthdata: CMR, CMR-STAC, TNM helper functions (use earthaccess instead)
io: reading and writing results to/from local files
ipysliderule: toolbox for building SlideRule interfaces in a Jupyter notebook

(2) Initialize the client#

sliderule.init("slideruleearth.io");

Configure the client settings:

url: address of sliderule service (default = “slideruleearth.io”)
verbose: display messages from server (default = False)
loglevel: criticality of log messages to display (default = logging.INFO)
organization: selection of cluster, used for private clusters (default = “sliderule”)
desired_nodes: number of nodes to run in a private cluster (default = None)
time_to_live: how long to deploy a private cluster (default = 60 minutes)
bypass_dns: query the provisioning system for IP address and don’t use DNS lookup hostname (default = False)
plugins: check if plugin is present (default = [])
trust_env: use netrc file for authentication (default = False)
log_handler: attach handler to client logging (default = None)
rethrow: immediately rethrow any caught exception inside of the client (default = None)

(3) Define an area of interest#

region = sliderule.toregion("grandmesa.geojson");

SlideRule uses an area of interest for determining which dataset resources to process and to then subset those resources to provide data only inside the area of interest. The sliderule.toregion function converts multiple input types into a format understood by SlideRule. The inputs types supported are: geojson, shapefile, GeoDataFrame, list of coordinates, and a dictionary of coordinates.

The resources (e.g. granules) to process can always be supplied in any of the processing APIs. But if they are not supplied (which is typical), then to determine which resources to process, the SlideRule server-side code uses the area of interest to make requests to NASA’s Common Metadata Repository (CMR) legacy and STAC interfaces, along with USGS’s The National Map interface. The server code automatically determines which interfaces should be queried and the parameters of the query needed for properly filtering results.

In rare cases when the area of interest is very complex (e.g. a bunch of islands, or an extremely high vertice-count polygon), then the user can request the server to rasterize the area of interest and use it as a mask for determining which data to process. See https://slideruleearth.io/web/rtd/user_guide/SlideRule.html#geojson for more details.

(4) Specify the processing parameters#

parms = {
    "poly": region["poly"],
    "srt": icesat2.SRT_LAND,
    "len": 20.0,
    "res": 100.0
}

There is a multitude of processing parameters that are available to each API. The ones used here are:

poly: area of interest
srt: surface reference type; if set to -1 (or icesat2.DYNAMIC), then all surface types are used
len: length of the extent (or variable-length segment) of along-track photon clouds to use in processing each posting
res: the step size between postings

See user’s guide for additional parameters: https://slideruleearth.io/web/rtd/index.html

(5) Make the processing request#

gdf = icesat2.atl06p(parms)

Under-the-hood this makes an HTTP request to the SlideRule service running in AWS to perform the ATL06 surface-finding algorithm on ATL03 photons to produce an elevation, and then collects the results into a pandas GeoDataFrame.

The different ICESat-2 APIs available are:

atl03sp: subset and filter ATL03 photons; provide custom YAPC and ATL08 classifications
atl03v: fast segment level subsetting of ATL03 photons
atl06s: subset the ATL06 land elevation product
atl06p: dynamically generate ATL06 surface elevation product
atl08p: dynamically generate the ATL08 vegetation density product (PhoREAL)
atl13p: subset the ATL13 coastal water product

Example 2: Sample GEDI Elevation Product at ICESat-2 Dynamically Generated Postings#

from sliderule import sliderule, icesat2, gedi

sliderule.init("slideruleearth.io", verbose=True);

Setting URL to slideruleearth.io
Login status to slideruleearth.io/sliderule: failure

parms = {
    "poly": sliderule.toregion('grandmesa.geojson')['poly'],
    "t0": '2019-11-14T00:00:00Z',
    "t1": '2019-11-15T00:00:00Z',
    "srt": icesat2.SRT_LAND,
    "len": 100,
    "res": 100,
    "pass_invalid": False, 
    "atl08_class": ["atl08_ground", "atl08_canopy", "atl08_top_of_canopy"],
    "atl08_fields": ["h_dif_ref"],
    "phoreal": {"binsize": 1.0, "geoloc": "center", "use_abs_h": False, "send_waveform": False},
    "samples": {"gedi": {"asset": "gedil3-elevation"}}
};

atl08 = icesat2.atl08p(parms)

request <AppServer.10153> retrieved 1 resources from CMR
proxy request <AppServer.10153> querying resources for gedi
proxy request <AppServer.10153> returned 0 resources for gedi
Starting proxy for atl08 to process 1 resource(s) with 1 thread(s)
request <AppServer.10171> processing initialized on ATL03_20191114034331_07370502_006_01.h5 ...
Successfully completed processing resource [1 out of 1]: ATL03_20191114034331_07370502_006_01.h5

atl08

	gt	h_min_canopy	rgt	veg_ph_count	landcover	x_atc	h_mean_canopy	segment_id	h_te_median	canopy_h_metrics	...	gnd_ph_count	h_max_canopy	cycle	canopy_openness	geometry	h_dif_ref	gedi.value	gedi.file_id	gedi.flags	gedi.time
time
2019-11-14 03:46:36.935118336	10	0.501465	737	27	30	-6.882999e+10	1.156670	215507	1958.305542	(1.479736328125, 1.479736328125, 1.47973632812...	...	75	2.261719	5	0.443189	POINT (-108.12262 38.83912)	0.963623	1777.066040	0	0	1.326586e+12
2019-11-14 03:46:36.949218304	10	0.515503	737	59	30	-6.882593e+10	1.304013	215512	1964.416748	(1.4913330078125, 1.4913330078125, 1.491333007...	...	54	3.137817	5	0.636816	POINT (-108.12272 38.84002)	-3.216064	1925.270142	0	0	1.326586e+12
2019-11-14 03:46:36.963318272	10	0.515259	737	54	30	-6.882186e+10	1.834195	215517	1976.178833	(1.5015869140625, 1.5015869140625, 1.501586914...	...	49	4.442627	5	0.892432	POINT (-108.12283 38.84092)	-5.043213	1925.270142	0	0	1.326586e+12
2019-11-14 03:46:36.977417984	10	0.804443	737	68	30	-6.881773e+10	2.465483	215522	1991.423218	(1.7940673828125, 1.7940673828125, 1.794067382...	...	35	6.167480	5	1.023235	POINT (-108.12295 38.84182)	2.426270	1925.270142	0	0	1.326586e+12
2019-11-14 03:46:36.980918016	30	0.508423	737	95	20	-6.927786e+10	2.816030	215529	1822.414673	(1.5048828125, 1.5048828125, 1.5048828125, 1.5...	...	19	7.769043	5	1.389679	POINT (-108.08651 38.84583)	5.912720	1779.992554	0	0	1.326586e+12
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
2019-11-14 03:46:42.283918336	60	0.520142	737	38	40	-6.851285e+10	2.928200	217296	1718.457764	(1.51220703125, 1.51220703125, 1.51220703125, ...	...	310	7.713501	5	2.058075	POINT (-108.08783 39.16624)	-0.977173	1781.542358	0	0	1.326586e+12
2019-11-14 03:46:42.293068288	20	0.504028	737	130	30	-6.805046e+10	3.273977	217288	1786.887939	(1.5029296875, 1.5029296875, 1.5029296875, 1.5...	...	192	8.225342	5	2.291607	POINT (-108.16125 39.1593)	-3.731567	1797.069336	0	0	1.326586e+12
2019-11-14 03:46:42.293818368	40	0.000000	737	0	40	-6.828026e+10	NaN	217294	1709.912354	(0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, 0.0, ...	...	272	0.000000	5	NaN	POINT (-108.12461 39.16313)	-0.086914	1720.139282	0	0	1.326586e+12
2019-11-14 03:46:42.295218176	60	0.548218	737	11	40	-6.851079e+10	4.089300	217301	1716.896362	(0.8890380859375, 0.8890380859375, 0.889038085...	...	191	10.701416	5	3.587460	POINT (-108.08792 39.16696)	-2.204712	1781.542358	0	0	1.326586e+12
2019-11-14 03:46:42.304318208	20	0.511963	737	13	20	-6.804797e+10	0.724375	217293	1784.938721	(1.41064453125, 1.41064453125, 1.41064453125, ...	...	161	1.410645	5	0.226650	POINT (-108.16134 39.16002)	-1.957886	1797.069336	0	0	1.326586e+12

2110 rows × 25 columns

Plot the results#

import matplotlib.pyplot as plt
import numpy as np

plt.figure(figsize=[8,6])

d0=np.min(atl08['x_atc'])

plt.plot(atl08['x_atc']-d0, atl08['h_te_median'], 'o',  markersize=1, color='green', label='h_mean_canopy')
plt.plot(atl08['x_atc']-d0, atl08['gedi.value'], 'o',  markersize=1, color='gray', label='gedi elevation')
hl=plt.legend(loc=3, frameon=False, markerscale=5)

plt.gca().set_ylim([1500, 3500])

(1500.0, 3500.0)

plt.show()

../../_images/63278f39e03470f3ce78a87255f76a28272cfaaf85d3e6abc996a0df157d7e85.png

Explanation of what’s new#

The on-demand ATL08 product (different than the ICESat-2 Standard Data Product) was generated and streamed back to the user. The ATL08 on-demand product uses University of Texas at Austin’s PhoREAL algorithm which was integrated into SlideRule to generate customizable vegetation metrics using ATL03 photon data.
A time range was specified in the request limiting the results to data collected only between the start and stop times supplied.
The "atl08_class" parameter specified that only photons in ATL03 that were classified as "atl08_ground", "atl08_canopy", or "atl08_top_of_canopy" in the ATL08 standard data product are to be supplied to the PhoREAL algorithm and used in the results.
The "atl08_fields" parameter specifies that the "h_dif_ref" variable from the ATL08 standard data product is to be associated with each result returned by SlideRule. SlideRule attempts to find the value of the variable closest in time to the dynamically generated result.
The "phoreal" parameter provides the processing parameters for the PhoREAL algorithm.
The "samples" parameter provides a list of raster datasets that SlideRule should sample at each generated result. So for each 100m segment that PhoREAL processes, the server-side code will also sample the gedil3-elevation product at the latitude and longitude of that segment and return the value with the results.

For a list of raster datasets that are available to sample in SlideRule, see: https://slideruleearth.io/web/rtd/user_guide/GeoRaster.html#asset-directory

Example 3: Produce GeoParquet of Coastal Photons#

from sliderule import sliderule, icesat2
import geopandas as gpd

sliderule.init(verbose=True)

Setting URL to slideruleearth.io
Login status to slideruleearth.io/sliderule: failure

True

region = sliderule.toregion("bathy.geojson");

# ATL03 subsetting request parameters
parms = {
    "poly": region['poly'],
    "srt": icesat2.SRT_DYNAMIC,
    "len": 100,
    "res": 100,
    "pass_invalid": True,
    "output": {
        "asset":"sliderule-stage",
        "format": "parquet",
        "as_geo": True,
        "open_on_complete": False
    }    
}

atl03_url = icesat2.atl03sp(parms, resources=['ATL03_20230213042035_08341807_006_02.h5'])

Starting proxy for atl03s to process 1 resource(s) with 1 thread(s)
request <AppServer.10175> processing initialized on ATL03_20230213042035_08341807_006_02.h5 ...
request <AppServer.10175> processing of ATL03_20230213042035_08341807_006_02.h5 complete (366784/0/0)
Initiated upload of results to S3, bucket = sliderule-public, key = sliderule.00000015394A5FCC.geoparquet
Upload to S3 completed, bucket = sliderule-public, key = sliderule.00000015394A5FCC.geoparquet, size = 10133778

atl03_url

's3://sliderule-public/sliderule.00000015394A5FCC.geoparquet'

# Recent issues with pandas and geopandas have made direct reads temperamental
# atl03 = gpd.pd.read_parquet(atl03_url)

import boto3
atl03_url_tokens = atl03_url.split('/')
s3_client = boto3.client('s3')
s3_client.download_file(atl03_url_tokens[2], atl03_url_tokens[3], "/tmp/" + atl03_url_tokens[3])

atl03 = gpd.read_parquet("/tmp/" + atl03_url_tokens[3])

atl03.keys()

Index(['extent_id', 'x_atc', 'landcover', 'y_atc', 'atl03_cnf', 'atl08_class',
       'snowcover', 'quality_ph', 'yapc_score', 'relief', 'height', 'cycle',
       'pair', 'sc_orient', 'rgt', 'track', 'background_rate', 'segment_id',
       'segment_dist', 'solar_elevation', 'region', 'geometry'],
      dtype='object')

Plot the results#

import matplotlib.pyplot as plt
import numpy as pd

df = atl03
df = df[df["pair"] == icesat2.LEFT_PAIR]
df = df[df["track"] == 3]
plt.figure(figsize=[8,6])
plt.plot(df['x_atc']+df['segment_dist'], df['height'], 'o',  markersize=1, color='blue')

[<matplotlib.lines.Line2D at 0x7f1e9f79b4d0>]

plt.show()

../../_images/5e1c20b1273cad6f9e4abfe69b29a4efc1068ddbad0c432904ab8057a4617b4b.png

Part 3: H5Coro - The HDF5 Cloud-Optimized Read-Only Python Package#

h5coro is a pure Python implementation of a subset of the HDF5 specification that has been optimized for reading data out of S3.

The project has its roots in SlideRule, where a new C++ implementation of the HDF5 specification was developed for performant read access to Earth science datasets stored in AWS S3. Over time, user’s of SlideRule began requesting the ability to performantly read HDF5 and NetCDF files out of S3 from their own Python scripts. The result is h5coro: the re-implementation in Python of the core HDF5 reading logic that exists in SlideRule. Since then, h5coro has become its own project, which will continue to grow and diverge in functionality from its parent implementation.

h5coro is optimized for reading HDF5 data in high-latency high-throughput environments. It accomplishes this through a few key design decisions:

All reads are concurrent. Each dataset and/or attribute read by h5coro is performed in its own thread.
Intelligent range gets are used to read as many dataset chunks as possible in each read operation. This drastically reduces the number of HTTP requests to S3 and means there is no longer a need to re-chunk the data (it actually works better on smaller chunk sizes due to the granularity of the request).
Block caching is used to minimize the number of GET requests made to S3. S3 has a large first-byte latency (we’ve measured it at ~60ms on our systems), which means there is a large penalty for each read operation performed. h5coro performs all reads to S3 as large block reads and then maintains data in a local cache for access to smaller amounts of data within those blocks.
The system is serverless and does not depend on any external services to read the data. This means it scales naturally as the user application scales, and it reduces overall system complexity.
No metadata repository is needed. The structure of the file are cached as they are read so that successive reads to other datasets in the same file will not have to re-read and re-build the directory structure of the file.

For more information:#

GitHub: SlideRuleEarth/h5coro

# To use the latest version of the sliderule client, run this cell.
# It will install the sliderule Python client into your current conda environment.
# You will then need to restart your kernel to have the changes take effect.
%pip install --quiet "h5coro>=0.0.7"

Note: you may need to restart the kernel to use updated packages.

Example 1: Read ATL03 variables for bathymetry#

# (1) Import modules
from h5coro import h5coro, s3driver
import earthaccess

# (2) Authenticate to Earth Data Login
auth = earthaccess.login()
s3_creds = auth.get_s3_credentials(daac="NSIDC")

# (3) Initialize h5coro object
granule = "nsidc-cumulus-prod-protected/ATLAS/ATL03/006/2023/02/13/ATL03_20230213042035_08341807_006_02.h5"
h5obj = h5coro.H5Coro(granule, s3driver.S3Driver, errorChecking=True, verbose=False, credentials=s3_creds, multiProcess=False)

# (4) Read the data
variables = ["/gt3l/heights/h_ph", "/gt3l/heights/dist_ph_along", "/gt3l/geolocation/segment_dist_x", "/gt3l/geolocation/segment_ph_cnt"]
promise = h5obj.readDatasets(variables, block=True, enableAttributes=False)
for variable in promise:
    print(f'{variable}: {promise[variable][0:10]}')

gt3l/heights/h_ph: [-47.941536 -51.9231   -48.09843  -47.873924 -48.12945  -48.118694
 -48.308052 -48.208042 -47.802708 -48.004234]
gt3l/heights/dist_ph_along: [0.7542868  0.76623714 1.4717534  2.187351   2.1880984  2.9048157
 2.905563   3.621905   4.337497   5.0545855 ]
gt3l/geolocation/segment_dist_x: [17068770.48934802 17068790.54479094 17068810.60023396 17068830.65567708
 17068850.7111203  17068870.76656362 17068890.82200704 17068910.87745056
 17068930.93289417 17068950.98833789]
gt3l/geolocation/segment_ph_cnt: [37 25 44 39 22 40 37 42 35 38]

Explanation of what happened#

(1) Import the necessary packages to use `h5coro`.#

h5coro relies on earthaccess for authenticating to Earth Data Login. The modules a user might want to import are:

s3driver: for reading data out of an s3 bucket
filedriver: for reading data out of a local file
webdriver: for reading data diretly over https (including objects in s3 buckets)
logger: for configuring the logging in h5coro

(3) Create an h5coro object for the granule that you want to read#

h5coro is object oriented, so all context information associated with the provided granule is stored in the object. Note that the full path to the granule is needed, including the s3 bucket.

(4) Read the data#

h5coro implements an asynchronous I/O interface, meaning that when the readDatasets function is called, it makes a read “request” in the background and returns immediately back to the caller. The caller receives something called a “promise” (or “future”) which is a promise that data will be there in the future at some point. You then can do other things while you wait, and when you finally need the data, you have to “block” or wait for it to be available.

In this example, I set the “block” parameter to True so that it would wait right away. But in more sophisticated examples, other work could have been done by the notebook while waiting for the results of the read.

Accessing ICESat-2 Data

Contents

Accessing ICESat-2 Data#

Part 1: icepyx#

Credit#

For more information#

Prerequisites#

Example 1: Search and Download ATL08 Granule#

Example 2: Reading a Granule with icepyx#

Create a `Read` object#

Explore your variables#

Load your variables#

Example 3: Reading a granule with h5py?#

Part 2: SlideRule#

For more information#

Example 1: Just Get Me Some Data#

Display the results#

Plot the results#

Explanation of what happened#

(1) Import the client#

(2) Initialize the client#

(3) Define an area of interest#

(4) Specify the processing parameters#

(5) Make the processing request#

Example 2: Sample GEDI Elevation Product at ICESat-2 Dynamically Generated Postings#

Plot the results#

Explanation of what’s new#

Example 3: Produce GeoParquet of Coastal Photons#

Plot the results#

Part 3: H5Coro - The HDF5 Cloud-Optimized Read-Only Python Package#

For more information:#

Example 1: Read ATL03 variables for bathymetry#

Explanation of what happened#

(1) Import the necessary packages to use `h5coro`.#

(3) Create an h5coro object for the granule that you want to read#

(4) Read the data#

Accessing ICESat-2 Data

Contents

Accessing ICESat-2 Data#

Part 1: icepyx#

Credit#

For more information#

Prerequisites#

Example 1: Search and Download ATL08 Granule#

Example 2: Reading a Granule with icepyx#

Create a Read object#

Explore your variables#

Load your variables#

Example 3: Reading a granule with h5py?#

Part 2: SlideRule#

For more information#

Example 1: Just Get Me Some Data#

Display the results#

Plot the results#

Explanation of what happened#

(1) Import the client#

(2) Initialize the client#

(3) Define an area of interest#

(4) Specify the processing parameters#

(5) Make the processing request#

Example 2: Sample GEDI Elevation Product at ICESat-2 Dynamically Generated Postings#

Plot the results#

Explanation of what’s new#

Example 3: Produce GeoParquet of Coastal Photons#

Plot the results#

Part 3: H5Coro - The HDF5 Cloud-Optimized Read-Only Python Package#

For more information:#

Example 1: Read ATL03 variables for bathymetry#

Explanation of what happened#

(1) Import the necessary packages to use h5coro.#

(2) Authenticate to Earth Data Login#

(3) Create an h5coro object for the granule that you want to read#

(4) Read the data#

Create a `Read` object#

(1) Import the necessary packages to use `h5coro`.#