Skip to content

Open in Colab

MUR SST L4

The Multi-scale Ultra-high Resolution (MUR) Sea Surface Temperature (SST) Level-4 products provide gap-free, globally gridded analyses by combining multiple satellite and in situ observations using interpolation and data fusion techniques. We will compare 2 MUR products. Both are on a regular grid:

  • MUR-JPL-L4-GLOB-v4.1 (~1 km resolution) High-resolution SST ~2002–present
  • MUR25-JPL-L4-GLOB-v04.2 (~25 km resolution) Coarse global SST analysis ~1992–2017 (retrospective)

Note: In a virtual machine in AWS us-west-2, where NASA cloud data is, the point matchups are fast. In Colab, say, your compute is not in the same data region nor provider, and the same matchups might take 10x longer.

Prerequisites

The examples here use NASA EarthData and you need to have an account with EarthData. Make sure you can login.

# if needed
!pip install point-collocation cartopy

import earthaccess
earthaccess.login()

Create some points

Random global over ocean.

import pandas as pd
url = (
    "https://raw.githubusercontent.com/"
    "fish-pace/point-collocation/main/"
    "examples/fixtures/points_1000.csv"
)
df_points = pd.read_csv(
    url,
    parse_dates=["time"]
)
df = df_points[
    (df_points["time"].dt.year == 2015) &
    (df_points["land"] == False)
]
print(len(df))
df.head()

17
lat lon time land
45 -16.216691 76.119033 2015-04-16 False
52 8.500092 82.677680 2015-07-21 False
251 5.963712 149.291272 2015-02-07 False
377 -10.447214 -155.929463 2015-10-30 False
384 -30.208351 10.304799 2015-01-12 False 
import matplotlib.pyplot as plt
import cartopy.crs as ccrs
import cartopy.feature as cfeature

# create Robinson projection
proj = ccrs.Robinson()

fig = plt.figure(figsize=(12,6))
ax = plt.axes(projection=proj)

# add map features
ax.add_feature(cfeature.LAND, facecolor="lightgray")
ax.add_feature(cfeature.OCEAN, facecolor="lightblue")
ax.add_feature(cfeature.COASTLINE, linewidth=0.5)

# plot points
ax.scatter(
    df["lon"],
    df["lat"],
    s=8,
    color="red",
    transform=ccrs.PlateCarree()
)

ax.set_global()
plt.title("Random Ocean Points (2015)")
plt.show()

png

Make a plan

import point_collocation as pc
short_name="MUR-JPL-L4-GLOB-v4.1"
plan = pc.plan(
    df,
    data_source="earthaccess",
    source_kwargs={
        "short_name": short_name,
    }
)
plan.summary(n=1)

Plan: 17 points → 17 unique granule(s)
  Points with 0 matches : 0
  Points with >1 matches: 0
  Time buffer: 0 days 00:00:00

First 1 point(s):
  [45] lat=-16.2167, lon=76.1190, time=2015-04-16 00:00:00: 1 match(es)
    → https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/MUR-JPL-L4-GLOB-v4.1/20150416090000-JPL-L4_GHRSST-SSTfnd-MUR-GLOB-v02.0-fv04.1.nc

Check out the data first

Hmm, very high resolution.

ds = plan.open_dataset(0)
ds

open_method: {'xarray_open': 'dataset', 'open_kwargs': {'chunks': {}, 'engine': 'h5netcdf', 'decode_timedelta': False}, 'coords': 'auto', 'set_coords': True, 'dim_renames': None, 'auto_align_phony_dims': None, 'merge': None}
Geolocation auto detected with cf_xarray: ('lon', 'lat') — lon dims=('lon',), lat dims=('lat',)

<xarray.Dataset> Size: 21GB
Dimensions:           (time: 1, lat: 17999, lon: 36000)
Coordinates:
  * time              (time) datetime64[ns] 8B 2015-01-12T09:00:00
  * lat               (lat) float32 72kB -89.99 -89.98 -89.97 ... 89.98 89.99
  * lon               (lon) float32 144kB -180.0 -180.0 -180.0 ... 180.0 180.0
Data variables:
    analysed_sst      (time, lat, lon) float64 5GB dask.array<chunksize=(1, 1023, 2047), meta=np.ndarray>
    analysis_error    (time, lat, lon) float64 5GB dask.array<chunksize=(1, 1023, 2047), meta=np.ndarray>
    mask              (time, lat, lon) float32 3GB dask.array<chunksize=(1, 1447, 2895), meta=np.ndarray>
    sea_ice_fraction  (time, lat, lon) float64 5GB dask.array<chunksize=(1, 1447, 2895), meta=np.ndarray>
    dt_1km_data       (time, lat, lon) float32 3GB dask.array<chunksize=(1, 1447, 2895), meta=np.ndarray>
Attributes: (12/47)
    Conventions:                CF-1.5
    title:                      Daily MUR SST, Final product
    summary:                    A merged, multi-sensor L4 Foundation SST anal...
    references:                 http://podaac.jpl.nasa.gov/Multi-scale_Ultra-...
    institution:                Jet Propulsion Laboratory
    history:                    created at nominal 4-day latency; replaced nr...
    ...                         ...
    project:                    NASA Making Earth Science Data Records for Us...
    publisher_name:             GHRSST Project Office
    publisher_url:              http://www.ghrsst.org
    publisher_email:            ghrsst-po@nceo.ac.uk
    processing_level:           L4
    cdm_data_type:              grid
# I am going to heavily coarsen before plotting
(
    ds.analysed_sst
    .coarsen(lat=100, lon=100, boundary="trim")
    .mean()
    .plot()
)

<matplotlib.collections.QuadMesh at 0x7f79d5aa13d0>

png

plan.open_dataset(0)

open_method: {'xarray_open': 'dataset', 'open_kwargs': {'chunks': {}, 'engine': 'h5netcdf', 'decode_timedelta': False}, 'coords': 'auto', 'set_coords': True, 'dim_renames': None, 'auto_align_phony_dims': None, 'merge': None}
Geolocation auto detected with cf_xarray: ('lon', 'lat') — lon dims=('lon',), lat dims=('lat',)

<xarray.Dataset> Size: 21GB
Dimensions:           (time: 1, lat: 17999, lon: 36000)
Coordinates:
  * time              (time) datetime64[ns] 8B 2015-01-12T09:00:00
  * lat               (lat) float32 72kB -89.99 -89.98 -89.97 ... 89.98 89.99
  * lon               (lon) float32 144kB -180.0 -180.0 -180.0 ... 180.0 180.0
Data variables:
    analysed_sst      (time, lat, lon) float64 5GB dask.array<chunksize=(1, 1023, 2047), meta=np.ndarray>
    analysis_error    (time, lat, lon) float64 5GB dask.array<chunksize=(1, 1023, 2047), meta=np.ndarray>
    mask              (time, lat, lon) float32 3GB dask.array<chunksize=(1, 1447, 2895), meta=np.ndarray>
    sea_ice_fraction  (time, lat, lon) float64 5GB dask.array<chunksize=(1, 1447, 2895), meta=np.ndarray>
    dt_1km_data       (time, lat, lon) float32 3GB dask.array<chunksize=(1, 1447, 2895), meta=np.ndarray>
Attributes: (12/47)
    Conventions:                CF-1.5
    title:                      Daily MUR SST, Final product
    summary:                    A merged, multi-sensor L4 Foundation SST anal...
    references:                 http://podaac.jpl.nasa.gov/Multi-scale_Ultra-...
    institution:                Jet Propulsion Laboratory
    history:                    created at nominal 4-day latency; replaced nr...
    ...                         ...
    project:                    NASA Making Earth Science Data Records for Us...
    publisher_name:             GHRSST Project Office
    publisher_url:              http://www.ghrsst.org
    publisher_email:            ghrsst-po@nceo.ac.uk
    processing_level:           L4
    cdm_data_type:              grid

Everything looks good

Let's try getting the point matchups. This would take about 5 hours on my machine for 10,000 points, but 30 seconds for our small dataset.

%%time
res = pc.matchup(plan, variables = ["sea_ice_fraction", 'analysed_sst', 'analysis_error'])
res

CPU times: user 7.08 s, sys: 862 ms, total: 7.94 s
Wall time: 30.5 s
lat lon time land pc_id granule_id granule_time granule_lat granule_lon sea_ice_fraction analysed_sst analysis_error
0 -16.216691 76.119033 2015-04-16 False 45 https://archive.podaac.earthdata.nasa.gov/poda... 2015-04-16 09:00:00+00:00 -16.219999 76.120003 NaN 301.205 0.39
1 8.500092 82.677680 2015-07-21 False 52 https://archive.podaac.earthdata.nasa.gov/poda... 2015-07-21 09:00:00+00:00 8.500000 82.680000 NaN 302.437 0.41
2 5.963712 149.291272 2015-02-07 False 251 https://archive.podaac.earthdata.nasa.gov/poda... 2015-02-07 09:00:00+00:00 5.960000 149.289993 NaN 303.088 0.41
3 -10.447214 -155.929463 2015-10-30 False 377 https://archive.podaac.earthdata.nasa.gov/poda... 2015-10-30 09:00:00+00:00 -10.450000 -155.929993 NaN 303.332 0.39
4 -30.208351 10.304799 2015-01-12 False 384 https://archive.podaac.earthdata.nasa.gov/poda... 2015-01-12 09:00:00+00:00 -30.209999 10.300000 NaN 294.948 0.37
5 42.798843 -46.120312 2015-09-15 False 396 https://archive.podaac.earthdata.nasa.gov/poda... 2015-09-15 09:00:00+00:00 42.799999 -46.119999 NaN 297.021 0.39
6 22.472809 170.643058 2015-04-09 False 439 https://archive.podaac.earthdata.nasa.gov/poda... 2015-04-09 09:00:00+00:00 22.469999 170.639999 NaN 298.274 0.38
7 -36.603864 118.546812 2015-03-18 False 448 https://archive.podaac.earthdata.nasa.gov/poda... 2015-03-18 09:00:00+00:00 -36.599998 118.550003 NaN 292.872 0.38
8 -10.106711 -24.478649 2015-04-25 False 457 https://archive.podaac.earthdata.nasa.gov/poda... 2015-04-25 09:00:00+00:00 -10.110000 -24.480000 NaN 301.137 0.38
9 -9.514659 143.858933 2015-01-20 False 596 https://archive.podaac.earthdata.nasa.gov/poda... 2015-01-20 09:00:00+00:00 -9.510000 143.860001 NaN 302.679 0.39
10 -19.014847 -128.909015 2015-12-22 False 622 https://archive.podaac.earthdata.nasa.gov/poda... 2015-12-22 09:00:00+00:00 -19.010000 -128.910004 NaN 301.250 0.38
11 42.296935 -70.544599 2015-02-15 False 660 https://archive.podaac.earthdata.nasa.gov/poda... 2015-02-15 09:00:00+00:00 42.299999 -70.540001 0.0 276.671 0.38
12 -25.067219 -154.574176 2015-07-03 False 818 https://archive.podaac.earthdata.nasa.gov/poda... 2015-07-03 09:00:00+00:00 -25.070000 -154.570007 NaN 295.126 0.37
13 48.835696 -2.650617 2015-01-30 False 851 https://archive.podaac.earthdata.nasa.gov/poda... 2015-01-30 09:00:00+00:00 48.840000 -2.650000 0.0 282.519 0.38
14 -53.953789 76.854571 2015-01-15 False 900 https://archive.podaac.earthdata.nasa.gov/poda... 2015-01-15 09:00:00+00:00 -53.950001 76.849998 0.0 273.502 0.38
15 -25.028446 95.599748 2015-02-13 False 994 https://archive.podaac.earthdata.nasa.gov/poda... 2015-02-13 09:00:00+00:00 -25.030001 95.599998 NaN 298.764 0.38
16 25.224213 60.241648 2015-11-11 False 999 https://archive.podaac.earthdata.nasa.gov/poda... 2015-11-11 09:00:00+00:00 25.219999 60.240002 NaN 301.702 0.38

Compare to the low-resolution product

This takes much less time per point since the files are smaller. This would take about an hour on my machine for 10,000 points.

import point_collocation as pc
short_name="MUR25-JPL-L4-GLOB-v04.2"
plan = pc.plan(
    df,
    data_source="earthaccess",
    source_kwargs={
        "short_name": short_name,
    }
)
plan.summary(n=1)

Plan: 17 points → 17 unique granule(s)
  Points with 0 matches : 0
  Points with >1 matches: 0
  Time buffer: 0 days 00:00:00

First 1 point(s):
  [45] lat=-16.2167, lon=76.1190, time=2015-04-16 00:00:00: 1 match(es)
    → https://archive.podaac.earthdata.nasa.gov/podaac-ops-cumulus-protected/MUR25-JPL-L4-GLOB-v04.2/20150416090000-JPL-L4_GHRSST-SSTfnd-MUR25-GLOB-v02.0-fv04.2.nc

plan.open_dataset(0)

open_method: {'xarray_open': 'dataset', 'open_kwargs': {'chunks': {}, 'engine': 'h5netcdf', 'decode_timedelta': False}, 'coords': 'auto', 'set_coords': True, 'dim_renames': None, 'auto_align_phony_dims': None, 'merge': None}
Geolocation auto detected with cf_xarray: ('lon', 'lat') — lon dims=('lon',), lat dims=('lat',)

<xarray.Dataset> Size: 37MB
Dimensions:           (time: 1, lat: 720, lon: 1440)
Coordinates:
  * time              (time) datetime64[ns] 8B 2015-01-12T09:00:00
  * lat               (lat) float32 3kB -89.88 -89.62 -89.38 ... 89.62 89.88
  * lon               (lon) float32 6kB -179.9 -179.6 -179.4 ... 179.6 179.9
Data variables:
    analysed_sst      (time, lat, lon) float64 8MB dask.array<chunksize=(1, 720, 1440), meta=np.ndarray>
    analysis_error    (time, lat, lon) float64 8MB dask.array<chunksize=(1, 720, 1440), meta=np.ndarray>
    mask              (time, lat, lon) float32 4MB dask.array<chunksize=(1, 720, 1440), meta=np.ndarray>
    sea_ice_fraction  (time, lat, lon) float64 8MB dask.array<chunksize=(1, 720, 1440), meta=np.ndarray>
    sst_anomaly       (time, lat, lon) float64 8MB dask.array<chunksize=(1, 720, 1440), meta=np.ndarray>
Attributes: (12/54)
    Conventions:                CF-1.7, ACDD-1.3
    title:                      Daily 0.25-degree MUR SST, Final product
    summary:                    A low-resolution version of the MUR SST analy...
    keywords:                   Oceans > Ocean Temperature > Sea Surface Temp...
    keywords_vocabulary:        NASA Global Change Master Directory (GCMD) Sc...
    standard_name_vocabulary:   NetCDF Climate and Forecast (CF) Metadata Con...
    ...                         ...
    publisher_name:             GHRSST Project Office
    publisher_url:              https://www.ghrsst.org
    publisher_email:            gpc@ghrsst.org
    file_quality_level:         3
    metadata_link:              http://podaac.jpl.nasa.gov/ws/metadata/datase...
    acknowledgment:             Please acknowledge the use of these data with...
%%time
res = pc.matchup(plan, variables = ["sea_ice_fraction", 'analysed_sst', 'analysis_error'])
res.head()

CPU times: user 4.31 s, sys: 16.3 ms, total: 4.33 s
Wall time: 6.09 s
lat lon time land pc_id granule_id granule_time granule_lat granule_lon sea_ice_fraction analysed_sst analysis_error
0 -16.216691 76.119033 2015-04-16 False 45 https://archive.podaac.earthdata.nasa.gov/poda... 2015-04-16 09:00:00+00:00 -16.125 76.125 NaN 301.427 0.39
1 8.500092 82.677680 2015-07-21 False 52 https://archive.podaac.earthdata.nasa.gov/poda... 2015-07-21 09:00:00+00:00 8.625 82.625 NaN 302.512 0.41
2 5.963712 149.291272 2015-02-07 False 251 https://archive.podaac.earthdata.nasa.gov/poda... 2015-02-07 09:00:00+00:00 5.875 149.375 NaN 303.106 0.41
3 -10.447214 -155.929463 2015-10-30 False 377 https://archive.podaac.earthdata.nasa.gov/poda... 2015-10-30 09:00:00+00:00 -10.375 -155.875 NaN 303.289 0.39
4 -30.208351 10.304799 2015-01-12 False 384 https://archive.podaac.earthdata.nasa.gov/poda... 2015-01-12 09:00:00+00:00 -30.125 10.375 NaN 294.973 0.37