Exploring PACE’s Role in Cyanobacteria Detection

Exploring PACE’s Role in Cyanobacteria Detection#

Author: Bingqing Liu (University of Louisiana Lafayette)

This tutorial will show you how to use HypterCoast to explore the spectra of a cyanobacteria bloom. We will use blooms in the Gulf of Mexico as an example.

What is HyperCoast?#

HyperCoast streamlines the processing of hyperspectral data from existing spaceborne and airborne missions (e.g., PACE, EMIT, AVIRIS, NEON, and DESIS) and upcoming hyperspectral missions, such as, SBG and GLIMR.

https://hypercoast.org

Uncomment the following cell to install the HyperCoast package.

# %pip install hypercoast
# %pip install -U hypercoast

Import libraries

import hypercoast
import pandas as pd

Read PACE level 2 data, which is non-grided nc file.

filepath = "../data/Bingqing/PACE_OCI.20240730T181157.L2.OC_AOP.V2_0.NRT.nc"
dataset = hypercoast.read_pace(filepath)
# Thanks to PACE OCI tutorials and notebooks: https://oceancolor.gsfc.nasa.gov/resources/docs/tutorials/notebooks/oci-file-structure/
plot = hypercoast.view_pace_pixel_locations(filepath, step=20)
dataset

<xarray.Dataset> Size: 2GB
Dimensions:     (latitude: 1710, longitude: 1272, wavelength: 184)
Coordinates:
    longitude   (latitude, longitude) float32 9MB ...
    latitude    (latitude, longitude) float32 9MB ...
  * wavelength  (wavelength) float64 1kB 339.0 341.0 344.0 ... 714.0 717.0 719.0
Data variables:
    Rrs         (latitude, longitude, wavelength) float32 2GB ...
Attributes:
    long_name:      Remote sensing reflectance
    units:          sr^-1
    standard_name:  surface_ratio_of_upwelling_radiance_emerging_from_sea_wat...
    valid_min:      -30000
    valid_max:      25000
../_images/41a687824b5ddd52e75c18dbba9742e89e0293be56437b7d75ca0d0e103aa09c.png

Load field data and interactively display it alongside PACE data.

kml_path = "../data/Bingqing/GOM_Hypoxia.kml"
filepath = "../data/Bingqing/Hypoxia_Data_Sheet.xlsx"
df = pd.read_excel(filepath)
df.head()
Station Station.1 Time Date Lon Lat Depth (m) Secchi (m) Salinity Water Temp ... Absorption CDOM LISST Nano Surface \npH Sufface \nO2 Bottom \nO2 FL-ECO FL-CDOM Notes
0 River stations NaN NaN NaT NaN NaN NaN NaN NaN NaN ... NaN NaN NaN NaN NaN NaN NaN NaN NaN NaN
1 R1 St1 09:39:00 2024-07-21 -89.45114 28.89887 NaN 2 28.71 30.74161 ... 125ml yes yes no NaN NaN NaN NaN NaN River Mouth
2 R2 St2 09:47:00 2024-07-21 -89.45306 28.90001 NaN 2 25.91325 30.91475 ... 150ml yes yes yes NaN NaN NaN NaN NaN River plume, seaside
3 R3 St3 09:59:00 2024-07-21 -89.43833 28.89487 NaN 0.75 24.44862 30.59565 ... 100ml yes yes yes NaN NaN NaN NaN NaN NaN
4 R4 St4 10:13:00 2024-07-21 -89.43162 28.90630 NaN 0.5 8.34838 30.13989 ... 100ml yes yes yes NaN NaN NaN NaN NaN NaN

5 rows × 22 columns

df_filtered = df.dropna(subset=['Lon', 'Lat']).reset_index(drop=True)
df_filtered.head()
Station Station.1 Time Date Lon Lat Depth (m) Secchi (m) Salinity Water Temp ... Absorption CDOM LISST Nano Surface \npH Sufface \nO2 Bottom \nO2 FL-ECO FL-CDOM Notes
0 R1 St1 09:39:00 2024-07-21 -89.45114 28.89887 NaN 2 28.71 30.74161 ... 125ml yes yes no NaN NaN NaN NaN NaN River Mouth
1 R2 St2 09:47:00 2024-07-21 -89.45306 28.90001 NaN 2 25.91325 30.91475 ... 150ml yes yes yes NaN NaN NaN NaN NaN River plume, seaside
2 R3 St3 09:59:00 2024-07-21 -89.43833 28.89487 NaN 0.75 24.44862 30.59565 ... 100ml yes yes yes NaN NaN NaN NaN NaN NaN
3 R4 St4 10:13:00 2024-07-21 -89.43162 28.90630 NaN 0.5 8.34838 30.13989 ... 100ml yes yes yes NaN NaN NaN NaN NaN NaN
4 R5 St5 10:58:00 2024-07-21 -89.37324 28.98095 NaN 0.5 2.4625 30.3054 ... 100ml yes yes yes NaN NaN NaN NaN NaN NaN

5 rows × 22 columns

Visualize PACE image and field data on an interactive map.

m = hypercoast.Map()
m.add_basemap("Hybrid")
wavelengths = [450, 550, 650]
m.add_pace(
    dataset, wavelengths, indexes=[3, 2, 1], vmin=0, vmax=0.02, layer_name="PACE"
)
m.add("spectral")
style = {"weight": 2, "color": "red"}
m.add_kml(kml_path, style=style, layer_name="Hypoxia Path", info_mode=None)
m.add_points_from_xy(df_filtered, x="Lon", y="Lat", max_cluster_radius=50, layer_name="Hypoxia Data Points")
m.set_center(-91.46118, 28.89758, zoom=8)
m

Checkout the demo:#

https://i.imgur.com/JdgotIT.gif

Compute band ratios.

da = dataset["Rrs"]
data = (da.sel(wavelength=650) > da.sel(wavelength=620)) & (da.sel(wavelength=701) > da.sel(wavelength=681)) & (da.sel(wavelength=701) > da.sel(wavelength=450))

The spectra of cyanobacteria bloom:#

Visualize the selected region based on band ratios.

import xarray as xr
import matplotlib.pyplot as plt
import cartopy.crs as ccrs
import cartopy.feature as cfeature

# Create a plot
fig, ax = plt.subplots(figsize=(12, 6), subplot_kw={'projection': ccrs.PlateCarree()})

#ax.set_extent([-93, -87, 28, 32], crs=ccrs.PlateCarree())

# Plot the data
data.plot(ax=ax, transform=ccrs.PlateCarree(), cmap='coolwarm', cbar_kwargs={'label': 'Cyano'},)

# Add coastlines
ax.coastlines()

# Add state boundaries
states_provinces = cfeature.NaturalEarthFeature(
    category='cultural',
    name='admin_1_states_provinces_lines',
    scale='50m',
    facecolor='none')

ax.add_feature(states_provinces, edgecolor='gray')

# Optionally, add gridlines, labels, etc.
ax.gridlines(draw_labels=True)

plt.show()
../_images/2838dabba383c30b06c02454dd3f3e0fb974289881eb29064b0ad1915977b343.png

What you can do next about Cyanobacteria using PACE?#

Spectral Angle Mapper: Spectral similarity Input: library of Cyanobacteria bloom Rrs spectra with Chla at different levels

Spectral Mixture Analysis: unmix different cyanobacteria species based on spectral difference.

Contents