MODIS MOD021KM and FIRMS - MODIS MOD021KM and FIRMS (Jupyter Notebook) published in the Environmental Data Science book

Context¶

Purpose¶

Explore MODIS satellite imagery and wildfire data that is open and free for scientific use.

Sensor description¶

The MOD021KM Giglio et al., 2018 product contains calibrated and geolocated at-aperture radiances for 36 discrete bands located in the 0.4 to 14.4 micron region of the electromagnetic spectrum.

Highlights¶

Use satpy to load, visualise, and regrid MODIS level 1B data.
Fetch a fire database containing some 497364 fires from 2020.
Visualisation of fire pixels from the database.
Visualisation of the fire pixels alongside bands from the MODIS satellite data.

Install and load libraries¶

Source

!pip -q install geopandas
!pip -q install geoviews

WARNING: The directory '/home/jovyan/.cache/pip' or its parent directory is not owned or is not writable by the current user. The cache has been disabled. Check the permissions and owner of that directory. If executing pip with sudo, you should use sudo's -H flag.

WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning.

WARNING: The directory '/home/jovyan/.cache/pip' or its parent directory is not owned or is not writable by the current user. The cache has been disabled. Check the permissions and owner of that directory. If executing pip with sudo, you should use sudo's -H flag.

WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning.

Source

import pandas as pd
import numpy as np
import geopandas
import intake
import fsspec, aiohttp
import hvplot.xarray
import hvplot.pandas
import holoviews as hv
import panel as pn
import satpy
import xarray as xr
import tempfile
from pathlib import Path
from scipy.spatial import cKDTree
from satpy.writers import get_enhanced_image
from getpass import getpass
from pathlib import Path
from pyresample import geometry
import datetime
import urllib.request
import os.path
from dotenv import load_dotenv
import warnings
warnings.filterwarnings(action='ignore')

Set project structure¶

notebook_folder = './notebook'
if not os.path.exists(notebook_folder):
    os.makedirs(notebook_folder)

Fetch Data¶

fname = 'MOD021KM.A2020245.0840.061.2020245193036.hdf'

if not os.path.isfile(os.path.join(notebook_folder, fname)) or os.path.getsize(os.path.join(notebook_folder, fname)) == 0:
    load_dotenv()
    username = os.environ.get("NASA_EARTHDATA_USER") #replace for your EarthData username if run local or in Binder
    password = os.environ.get("NASA_EARTHDATA_PWD")  #replace for your EarthData password if run local or in Binder

    fsspec.config.conf['https'] = dict(client_kwargs={'auth': aiohttp.BasicAuth(username, password)})

    url = f'https://ladsweb.modaps.eosdis.nasa.gov/archive/allData/61/MOD021KM/2020/245/{fname}'
    filename = url.split('/')[-1]
    with fsspec.open(url) as f:
        with Path(os.path.join(notebook_folder, filename)).open('wb') as handle:
            data = f.read()
            try:
                data.decode('utf-8')
                raise RuntimeError('Could not download MODIS data! Have you authorized LAADS Web in your Eathdata account above?')
            except UnicodeDecodeError:
                handle.write(data)

Download the database of MODIS wildfires from a publically accessible location.

filepath = 'https://firms2.modaps.eosdis.nasa.gov/data/country/zips'
filename = 'modis_2020_all_countries.zip'

if not os.path.isfile(filename) or os.path.getsize(filename) == 0:
    urllib.request.urlretrieve(os.path.join(filepath, filename), os.path.join(notebook_folder, filename))
    !unzip -o ./notebook/modis_2020_all_countries.zip -d ./notebook

Archive:  ./notebook/modis_2020_all_countries.zip
   creating: ./notebook/modis/2020/
  inflating: ./notebook/modis/2020/modis_2020_Afghanistan.csv  
  inflating: ./notebook/modis/2020/modis_2020_Albania.csv  
  inflating: ./notebook/modis/2020/modis_2020_Algeria.csv  
  inflating: ./notebook/modis/2020/modis_2020_Andorra.csv  
  inflating: ./notebook/modis/2020/modis_2020_Angola.csv


  inflating: ./notebook/modis/2020/modis_2020_Antigua_and_Barbuda.csv  
  inflating: ./notebook/modis/2020/modis_2020_Argentina.csv

  inflating: ./notebook/modis/2020/modis_2020_Armenia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Australia.csv


  inflating: ./notebook/modis/2020/modis_2020_Austria.csv  
  inflating: ./notebook/modis/2020/modis_2020_Azerbaijan.csv  
  inflating: ./notebook/modis/2020/modis_2020_Bahamas.csv  
  inflating: ./notebook/modis/2020/modis_2020_Bahrain.csv  
  inflating: ./notebook/modis/2020/modis_2020_Bangladesh.csv  
  inflating: ./notebook/modis/2020/modis_2020_Barbados.csv  
  inflating: ./notebook/modis/2020/modis_2020_Belarus.csv  
  inflating: ./notebook/modis/2020/modis_2020_Belgium.csv  
  inflating: ./notebook/modis/2020/modis_2020_Belize.csv  
  inflating: ./notebook/modis/2020/modis_2020_Benin.csv  
  inflating: ./notebook/modis/2020/modis_2020_Bermuda.csv  
  inflating: ./notebook/modis/2020/modis_2020_Bhutan.csv  
  inflating: ./notebook/modis/2020/modis_2020_Bolivia.csv


  inflating: ./notebook/modis/2020/modis_2020_Bosnia_and_Herzegovina.csv  
  inflating: ./notebook/modis/2020/modis_2020_Botswana.csv  
  inflating: ./notebook/modis/2020/modis_2020_Brazil.csv


  inflating: ./notebook/modis/2020/modis_2020_British_Virgin_Islands.csv  
  inflating: ./notebook/modis/2020/modis_2020_Brunei_Darussalam.csv  
  inflating: ./notebook/modis/2020/modis_2020_Bulgaria.csv  
  inflating: ./notebook/modis/2020/modis_2020_Burkina_Faso.csv  
  inflating: ./notebook/modis/2020/modis_2020_Burundi.csv  
  inflating: ./notebook/modis/2020/modis_2020_Cambodia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Cameroon.csv


  inflating: ./notebook/modis/2020/modis_2020_Canada.csv  
  inflating: ./notebook/modis/2020/modis_2020_Cape_Verde.csv  
  inflating: ./notebook/modis/2020/modis_2020_Cayman_Islands.csv  
  inflating: ./notebook/modis/2020/modis_2020_Central_African_Republic.csv


  inflating: ./notebook/modis/2020/modis_2020_Chad.csv  
  inflating: ./notebook/modis/2020/modis_2020_Chile.csv  
  inflating: ./notebook/modis/2020/modis_2020_China.csv


  inflating: ./notebook/modis/2020/modis_2020_Colombia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Comoros.csv  
  inflating: ./notebook/modis/2020/modis_2020_Costa_Rica.csv  
  inflating: ./notebook/modis/2020/modis_2020_Cote_d_Ivoire.csv  
  inflating: ./notebook/modis/2020/modis_2020_Croatia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Cuba.csv


  inflating: ./notebook/modis/2020/modis_2020_Curacao.csv  
  inflating: ./notebook/modis/2020/modis_2020_Cyprus.csv  
  inflating: ./notebook/modis/2020/modis_2020_Czech_Republic.csv  
  inflating: ./notebook/modis/2020/modis_2020_Dem_Rep_Korea.csv  
  inflating: ./notebook/modis/2020/modis_2020_Democratic_Republic_of_the_Congo.csv


  inflating: ./notebook/modis/2020/modis_2020_Denmark.csv  
  inflating: ./notebook/modis/2020/modis_2020_Djibouti.csv  
  inflating: ./notebook/modis/2020/modis_2020_Dominica.csv  
  inflating: ./notebook/modis/2020/modis_2020_Dominican_Republic.csv  
  inflating: ./notebook/modis/2020/modis_2020_Ecuador.csv  
  inflating: ./notebook/modis/2020/modis_2020_Egypt.csv  
  inflating: ./notebook/modis/2020/modis_2020_El_Salvador.csv  
  inflating: ./notebook/modis/2020/modis_2020_Equatorial_Guinea.csv  
  inflating: ./notebook/modis/2020/modis_2020_Eritrea.csv  
  inflating: ./notebook/modis/2020/modis_2020_Estonia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Ethiopia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Falkland_Islands.csv  
  inflating: ./notebook/modis/2020/modis_2020_Fiji.csv  
  inflating: ./notebook/modis/2020/modis_2020_Finland.csv  
  inflating: ./notebook/modis/2020/modis_2020_France.csv  
  inflating: ./notebook/modis/2020/modis_2020_French_Guiana.csv  
  inflating: ./notebook/modis/2020/modis_2020_French_Polynesia.csv  
  inflating: ./notebook/modis/2020/modis_2020_French_Southern_and_Antarctic_Lands.csv  
  inflating: ./notebook/modis/2020/modis_2020_Gabon.csv  
  inflating: ./notebook/modis/2020/modis_2020_Georgia.csv

  inflating: ./notebook/modis/2020/modis_2020_Germany.csv  
  inflating: ./notebook/modis/2020/modis_2020_Ghana.csv  
  inflating: ./notebook/modis/2020/modis_2020_Greece.csv  
  inflating: ./notebook/modis/2020/modis_2020_Greenland.csv  
  inflating: ./notebook/modis/2020/modis_2020_Guadeloupe.csv  
  inflating: ./notebook/modis/2020/modis_2020_Guam.csv  
  inflating: ./notebook/modis/2020/modis_2020_Guatemala.csv  
  inflating: ./notebook/modis/2020/modis_2020_Guinea-Bissau.csv  
  inflating: ./notebook/modis/2020/modis_2020_Guinea.csv


  inflating: ./notebook/modis/2020/modis_2020_Guyana.csv  
  inflating: ./notebook/modis/2020/modis_2020_Haiti.csv  
  inflating: ./notebook/modis/2020/modis_2020_Heard_I_and_McDonald_Islands.csv  
  inflating: ./notebook/modis/2020/modis_2020_Honduras.csv  
  inflating: ./notebook/modis/2020/modis_2020_Hong_Kong.csv  
  inflating: ./notebook/modis/2020/modis_2020_Hungary.csv  
  inflating: ./notebook/modis/2020/modis_2020_India.csv


  inflating: ./notebook/modis/2020/modis_2020_Indonesia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Iran.csv  
  inflating: ./notebook/modis/2020/modis_2020_Iraq.csv  
  inflating: ./notebook/modis/2020/modis_2020_Ireland.csv  
  inflating: ./notebook/modis/2020/modis_2020_Israel.csv  
  inflating: ./notebook/modis/2020/modis_2020_Italy.csv


  inflating: ./notebook/modis/2020/modis_2020_Jamaica.csv  
  inflating: ./notebook/modis/2020/modis_2020_Japan.csv  
  inflating: ./notebook/modis/2020/modis_2020_Jordan.csv  
  inflating: ./notebook/modis/2020/modis_2020_Kazakhstan.csv  
  inflating: ./notebook/modis/2020/modis_2020_Kenya.csv  
  inflating: ./notebook/modis/2020/modis_2020_Kiribati.csv  
  inflating: ./notebook/modis/2020/modis_2020_Kosovo.csv  
  inflating: ./notebook/modis/2020/modis_2020_Kuwait.csv  
  inflating: ./notebook/modis/2020/modis_2020_Kyrgyzstan.csv  
  inflating: ./notebook/modis/2020/modis_2020_Lao_PDR.csv  
  inflating: ./notebook/modis/2020/modis_2020_Latvia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Lebanon.csv  
  inflating: ./notebook/modis/2020/modis_2020_Lesotho.csv  
  inflating: ./notebook/modis/2020/modis_2020_Liberia.csv


  inflating: ./notebook/modis/2020/modis_2020_Libya.csv  
  inflating: ./notebook/modis/2020/modis_2020_Lithuania.csv  
  inflating: ./notebook/modis/2020/modis_2020_Luxembourg.csv  
  inflating: ./notebook/modis/2020/modis_2020_Macedonia_Former_Yugoslav_Republic_of.csv  
  inflating: ./notebook/modis/2020/modis_2020_Madagascar.csv  
  inflating: ./notebook/modis/2020/modis_2020_Malawi.csv  
  inflating: ./notebook/modis/2020/modis_2020_Malaysia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Maldives.csv  
  inflating: ./notebook/modis/2020/modis_2020_Mali.csv


  inflating: ./notebook/modis/2020/modis_2020_Malta.csv  
  inflating: ./notebook/modis/2020/modis_2020_Martinique.csv  
  inflating: ./notebook/modis/2020/modis_2020_Mauritania.csv  
  inflating: ./notebook/modis/2020/modis_2020_Mauritius.csv  
  inflating: ./notebook/modis/2020/modis_2020_Mexico.csv  
  inflating: ./notebook/modis/2020/modis_2020_Moldova.csv  
  inflating: ./notebook/modis/2020/modis_2020_Mongolia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Montenegro.csv  
  inflating: ./notebook/modis/2020/modis_2020_Montserrat.csv  
  inflating: ./notebook/modis/2020/modis_2020_Morocco.csv  
  inflating: ./notebook/modis/2020/modis_2020_Mozambique.csv


  inflating: ./notebook/modis/2020/modis_2020_Myanmar.csv


  inflating: ./notebook/modis/2020/modis_2020_Namibia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Nepal.csv  
  inflating: ./notebook/modis/2020/modis_2020_Netherlands.csv  
  inflating: ./notebook/modis/2020/modis_2020_New_Caledonia.csv  
  inflating: ./notebook/modis/2020/modis_2020_New_Zealand.csv  
  inflating: ./notebook/modis/2020/modis_2020_Nicaragua.csv  
  inflating: ./notebook/modis/2020/modis_2020_Niger.csv  
  inflating: ./notebook/modis/2020/modis_2020_Nigeria.csv


  inflating: ./notebook/modis/2020/modis_2020_Northern_Mariana_Islands.csv  
  inflating: ./notebook/modis/2020/modis_2020_Norway.csv  
  inflating: ./notebook/modis/2020/modis_2020_Oman.csv  
  inflating: ./notebook/modis/2020/modis_2020_Pakistan.csv  
  inflating: ./notebook/modis/2020/modis_2020_Palau.csv  
  inflating: ./notebook/modis/2020/modis_2020_Palestine.csv  
  inflating: ./notebook/modis/2020/modis_2020_Panama.csv  
  inflating: ./notebook/modis/2020/modis_2020_Papua_New_Guinea.csv  
  inflating: ./notebook/modis/2020/modis_2020_Paraguay.csv


  inflating: ./notebook/modis/2020/modis_2020_Peru.csv  
  inflating: ./notebook/modis/2020/modis_2020_Philippines.csv  
  inflating: ./notebook/modis/2020/modis_2020_Poland.csv  
  inflating: ./notebook/modis/2020/modis_2020_Portugal.csv  
  inflating: ./notebook/modis/2020/modis_2020_Puerto_Rico.csv  
  inflating: ./notebook/modis/2020/modis_2020_Qatar.csv  
  inflating: ./notebook/modis/2020/modis_2020_Republic_of_Congo.csv  
  inflating: ./notebook/modis/2020/modis_2020_Republic_of_Korea.csv  
  inflating: ./notebook/modis/2020/modis_2020_Reunion.csv  
  inflating: ./notebook/modis/2020/modis_2020_Romania.csv  
  inflating: ./notebook/modis/2020/modis_2020_Russian_Federation.csv


  inflating: ./notebook/modis/2020/modis_2020_Rwanda.csv  
  inflating: ./notebook/modis/2020/modis_2020_Saint_Helena.csv  
  inflating: ./notebook/modis/2020/modis_2020_Saint_Kitts_and_Nevis.csv  
  inflating: ./notebook/modis/2020/modis_2020_Saint_Lucia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Saint_Vincent_and_the_Grenadines.csv  
  inflating: ./notebook/modis/2020/modis_2020_Samoa.csv  
  inflating: ./notebook/modis/2020/modis_2020_Sao_Tome_and_Principe.csv  
  inflating: ./notebook/modis/2020/modis_2020_Saudi_Arabia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Senegal.csv  
  inflating: ./notebook/modis/2020/modis_2020_Serbia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Seychelles.csv  
  inflating: ./notebook/modis/2020/modis_2020_Sierra_Leone.csv  
  inflating: ./notebook/modis/2020/modis_2020_Singapore.csv  
  inflating: ./notebook/modis/2020/modis_2020_Slovakia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Slovenia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Solomon_Islands.csv  
  inflating: ./notebook/modis/2020/modis_2020_Somalia.csv  
  inflating: ./notebook/modis/2020/AttributesReadme.txt  
  inflating: ./notebook/modis/2020/modis_2020_South_Africa.csv  
  inflating: ./notebook/modis/2020/modis_2020_South_Sudan.csv


  inflating: ./notebook/modis/2020/modis_2020_Spain.csv  
  inflating: ./notebook/modis/2020/modis_2020_Sri_Lanka.csv  
  inflating: ./notebook/modis/2020/modis_2020_Sudan.csv


  inflating: ./notebook/modis/2020/modis_2020_Suriname.csv  
  inflating: ./notebook/modis/2020/modis_2020_Svalbard_and_Jan_Mayen.csv  
  inflating: ./notebook/modis/2020/modis_2020_Swaziland.csv  
  inflating: ./notebook/modis/2020/modis_2020_Sweden.csv  
  inflating: ./notebook/modis/2020/modis_2020_Switzerland.csv  
  inflating: ./notebook/modis/2020/modis_2020_Syria.csv  
  inflating: ./notebook/modis/2020/modis_2020_Taiwan.csv  
  inflating: ./notebook/modis/2020/modis_2020_Tajikistan.csv  
  inflating: ./notebook/modis/2020/modis_2020_Tanzania.csv  
  inflating: ./notebook/modis/2020/modis_2020_Thailand.csv


  inflating: ./notebook/modis/2020/modis_2020_The_Gambia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Timor-Leste.csv  
  inflating: ./notebook/modis/2020/modis_2020_Togo.csv  
  inflating: ./notebook/modis/2020/modis_2020_Tonga.csv  
  inflating: ./notebook/modis/2020/modis_2020_Trinidad_and_Tobago.csv  
  inflating: ./notebook/modis/2020/modis_2020_Tunisia.csv  
  inflating: ./notebook/modis/2020/modis_2020_Turkey.csv  
  inflating: ./notebook/modis/2020/modis_2020_Turkmenistan.csv  
  inflating: ./notebook/modis/2020/modis_2020_Turks_and_Caicos_Islands.csv  
  inflating: ./notebook/modis/2020/modis_2020_Uganda.csv  
  inflating: ./notebook/modis/2020/modis_2020_Ukraine.csv  
  inflating: ./notebook/modis/2020/modis_2020_United_Arab_Emirates.csv  
  inflating: ./notebook/modis/2020/modis_2020_United_Kingdom.csv  
  inflating: ./notebook/modis/2020/modis_2020_United_States.csv


  inflating: ./notebook/modis/2020/modis_2020_United_States_Minor_Outlying_Islands.csv  
  inflating: ./notebook/modis/2020/modis_2020_United_States_Virgin_Islands.csv  
  inflating: ./notebook/modis/2020/modis_2020_Uruguay.csv  
  inflating: ./notebook/modis/2020/modis_2020_Uzbekistan.csv  
  inflating: ./notebook/modis/2020/modis_2020_Vanuatu.csv  
  inflating: ./notebook/modis/2020/modis_2020_Venezuela.csv


  inflating: ./notebook/modis/2020/modis_2020_Vietnam.csv  
  inflating: ./notebook/modis/2020/modis_2020_Yemen.csv  
  inflating: ./notebook/modis/2020/modis_2020_Zambia.csv


  inflating: ./notebook/modis/2020/modis_2020_Zimbabwe.csv

Load an intake catalog for the downloaded data

# set catalogue location
catalog_file = os.path.join(notebook_folder, 'catalog.yaml')

with open(catalog_file, 'w') as f:
    f.write('''
sources:
    modis_l1b:
      args:
        urlpath: 'notebook/MOD021KM.A2020245.0840.061.2020245193036.hdf'
        reader: 'modis_l1b'
      description: "MODIS Level 1B Products"
      driver: SatpySource
      metadata: {}
    modis_fires:
      args:
        urlpath: 'notebook/modis/2020/modis_*.csv'
      description: "MODIS Level 2 Fires"
      driver: csv
      metadata: {}
''')

from intake.source.base import PatternMixin
from intake.source.base import DataSource, Schema

import glob

class SatpySource(DataSource, PatternMixin):
    """Intake driver for data supported by satpy.Scene"""
    
    container = 'python'
    name = 'satpy'
    version = '0.0.1'
    partition_access = False

    def __init__(self, urlpath, reader=None, metadata=None, path_as_pattern=True):
        """
        Parameters
        ----------
        path: str, location of model pkl file
        Either the absolute or relative path to the file
        opened. Some examples:
          - ``{{ CATALOG_DIR }}/file.nat``
          - ``{{ CATALOG_DIR }}/*.nc``
        reader: str, optional
        Name of the satpy reader to use when loading data (ie. ``modis_l1b``)
        metadata: dict, optional
        Additional metadata to pass to the intake catalog
        path_as_pattern: bool or str, optional
        Whether to treat the path as a pattern (ie. ``data_{field}.nc``)
        and create new coodinates in the output corresponding to pattern
        fields. If str, is treated as pattern to match on. Default is True.
        """

        self.path_as_pattern = path_as_pattern
        self.urlpath = urlpath
        self._reader = reader
        super(SatpySource, self).__init__(metadata=metadata)

    def _load(self):
        files = self.urlpath
        files = list(glob.glob(files))
        return satpy.Scene(files, reader=self._reader)
    
    def _get_schema(self):
        self._schema = Schema(
            npartitions=1,
            extra_metadata={}
        )
        return self._schema

    def read(self):
        self._load_metadata()
        return self._load()

intake.register_driver('SatpySource', SatpySource)
cat = intake.open_catalog(catalog_file)

Load MODIS Satellite Data¶

Here we use satpy to load the MODIS level 1b data into memory. As well as loading the image data, satpy provides a easy way to regrid the data to different coordinate systems.

scn = cat['modis_l1b'].read()
scn

<satpy.scene.Scene at 0x7f7bf832efa0>

Source

scn.load(['true_color', '20'], resolution=1000)
plot_modis_raw = scn.show('true_color')
plot_modis_raw

  0%|          | 0/52 [00:00<?, ?kB/s]

100%|██████████| 52/52 [00:00<00:00, 1296.98kB/s]

No rsr file /home/jovyan/.local/share/pyspectral/rsr_modis_EOS-Terra.h5 on disk

  0%|          | 0/5 [00:00<?, ?kB/s]

100%|██████████| 5/5 [00:00<00:00, 1054.91kB/s]

<trollimage.xrimage.XRImage at 0x7f7bf31080a0>

Resample to a different projection¶

In the plot above you can see that the raw MODIS data has distortion towards edge of the image. By regridding the data we can avoid some of this distortion.

area_id = "Southern Africa"
description = "Southern Africa in Mercator-Projection"
proj_id = "Southern Africa"
proj_dict = {"proj": "eqc"}

width = 1000    # width of the result domain in pixels
height = 1000   # height of the result domain in pixels

llx =  23E5   # projection x coordinate of lower left corner of lower left pixel
lly =  -22.9E5   # projection y coordinate of lower left corner of lower left pixel
urx =  48E5   # projection x coordinate of upper right corner of upper right pixel
ury =  -1.9E5   # projection y coordinate of upper right corner of upper right pixel

area_extent = (llx,lly,urx,ury)
from pyresample import create_area_def
resolution=1000
center =(0,0)
area_def = create_area_def(area_id, proj_dict, center=center, resolution=resolution)

modis_scn = scn.resample(area_def, radius_of_influence=10000)

Source

plot_modis_scn = modis_scn.show('true_color')
plot_modis_scn

<trollimage.xrimage.XRImage at 0x7f7bf8386ca0>

Load MODIS Fire Database¶

Now we’re going to load the modis fire database from CSV files. This contains the longitude, latitude, and time of where fires have been detected to occur. It also includes an estimate of the Fire Radiative Power (FRP), a measure of the intensity of the fire, for each fire detected.

Source

fires = cat['modis_fires'].read()

time = fires.acq_date.astype(str) + ' ' + fires.acq_time.astype(str).str.zfill(4)
fires['timestamp'] = pd.to_datetime(time, format='%Y-%m-%d %H%M')
fires = fires.loc[fires.satellite == 'Terra']
fires = geopandas.GeoDataFrame(
    fires, geometry=geopandas.points_from_xy(fires.longitude, fires.latitude))

# We're only interested in data from Southern Africa for now
llx =  0    # projection x coordinate of lower left corner of lower left pixel
lly =  -30  # projection y coordinate of lower left corner of lower left pixel
urx =  55   # projection x coordinate of upper right corner of upper right pixel
ury =  10   # projection y coordinate of upper right corner of upper right pixel

fires = fires.cx[llx:urx, lly:ury]
fires.set_index('timestamp', drop=False, inplace=True)
fires = fires.sort_index()
fires = fires.loc['2020-09-01 00:00:00':'2020-10-01 00:00:00']
fires

Daily Fire Radiative Power¶

Here we plot the Fire Radiative Power (FRP) for each day of the month of September. FRP is a measure of the intensity of a fire in units of MegaWatts.

Source

from bokeh.models.formatters import DatetimeTickFormatter

def plot_timeseries(data, y_variable):
    """Timeseries plot showing the mean fire radiative power at each day as well as the 5% and 95%"""

    def percentile(n):
        def percentile_(x):
            return np.percentile(x, n)
        percentile_.__name__ = 'percentile_%s' % n
        return percentile_

    formatter = DatetimeTickFormatter(months='%b')

    bounds = data.groupby([data.index.day_of_year])[y_variable].agg([percentile(5), percentile(95)])
    avg = data.groupby([data.index.day_of_year])[y_variable].mean()

    bounds.index = data.groupby([data.index.day_of_year])['timestamp'].first()
    avg.index = bounds.index

    tseries = hv.Overlay([
        (bounds.hvplot.area('timestamp', 'percentile_5', 'percentile_95',
                      alpha=0.2, color='red', xformatter=formatter)),
            avg.hvplot.line(x='timestamp', label=f'Average Daily FRP', color='red', xformatter=formatter)])

    return tseries.options(width=800, height=400, xlabel='Day',ylabel=f'FRP (MW)',legend_position='top_left')

plot_FIRMS_ts = plot_timeseries(fires, 'frp')
plot_FIRMS_ts

Geographical distribution of Fires¶

Here we plot the geographical distribution of fires detected by MODIS over Southern Africa.

Source

plot_FIRMS_geo = fires.hvplot.points('longitude', 'latitude', groupby='index.day_of_year', c='frp', geo=True, alpha=0.2,
                    tiles='ESRI', xlim=(llx, urx), ylim=(lly, ury), cmap='autumn', logz=True,
                    dynamic=False)
plot_FIRMS_geo

Visualising Fire Pixels with Satellite Imagery¶

Visualise a color image of the MODIS region using hvplot.

Source

modis_dataset = modis_scn.to_xarray_dataset()

img = get_enhanced_image(modis_scn['true_color'].squeeze())
img = img.data
img = img.clip(0, 1)
img = (img * 255).astype(np.uint8)

modis_dataset['true_color'] = img

grid = modis_scn.min_area().get_lonlats()
lons, lats = grid[0][0], grid[1][:, 0]
modis_dataset = modis_dataset.assign_coords(dict(x=lons, y=lats))
modis_dataset = modis_dataset.rename(dict(x='longitude', y='latitude'))

plot_MODIS_rgb = modis_dataset['true_color'].hvplot.rgb(x='longitude', y='latitude', bands='bands', rasterize=True)
plot_MODIS_rgb

Get the fire pixels that are visible in the the MODIS scene by filtering the time, longitude & latitude. We can also compute which row and column the fire pixel is in for this projection.

Source

grid = modis_scn.min_area().get_lonlats()

# Mask any fire pixels not in this scene
time_mask = np.logical_and(fires.timestamp >= scn.start_time,
                           fires.timestamp <= scn.end_time)
fire_points = fires.loc[time_mask]
fire_points = fire_points.cx[grid[0].min():grid[0].max(), grid[1].min():grid[1].max()]

# extract the x and y coordinates as flat arrays
x = np.ravel(grid[0])
y = np.ravel(grid[1])
points = np.dstack([x, y]).squeeze()

# Find locations of fire pixels within satellite image
tree = cKDTree(points)
distance, idx = tree.query(fire_points[['longitude', 'latitude']], k=1)
index = np.unravel_index(idx, grid[0].shape)
index = np.vstack(index).T
index

# fire_points[['y', 'x']] = index
fire_points = fire_points[img.values[0, index[:, 0], index[:, 1]] != 0]
fire_points

Now overlay the fire pixels ontop of the MODIS image, along with the FRP for each fire pixel. Try zooming in, you should be able to see clear smoke trails at the locations of some of the fires!

Source

rgb = modis_dataset['true_color'].hvplot.rgb(x='longitude', y='latitude', bands='bands', data_aspect=1, hover=False, title='True Colour', rasterize=True)
points = fire_points.hvplot.points('longitude', 'latitude', c='frp', cmap='autumn', logz=True, alpha=0.4)
plot_FIRMS_MODIS_rgb = rgb*points
plot_FIRMS_MODIS_rgb

We can also overlay the fire pixels ontop of the MODIS 3.7 micron wavelength band. The 3.7 band is a thermal band. Fires will appear as very bright pixels in image. Try zooming in, you should be able to clearly see bright spots at the location of each fire pixel.

Source

thermal = modis_dataset['20'].hvplot.image(x='longitude', y='latitude', cmap='viridis', data_aspect=1, hover=False, title='Band 20: 3.7 micron', rasterize=True)
points = fire_points.hvplot.points('longitude', 'latitude', c='frp', cmap='autumn', logz=True, alpha=0.4)
plot_FIRMS_MODIS_thermal = thermal*points
plot_FIRMS_MODIS_thermal

Summary¶

This notebook has demonstrated the use of:

The satpy package to easily load, plot and regrid satellite data from the MODIS sensor.
hvplot to interactively visualise wildfire pixels detected from the MODIS sensor.
geopandas to load, filter, and manipulate historical wildfire pixel data.

Citing this Notebook¶

Please see CITATION.cff for the full citation information. The citation file can be exported to APA or BibTex formats (learn more here).

Additional information¶

Review: This notebook has been reviewed by one or more members of the Environmental Data Science book community. The open review is available here.

Dataset: MOD021KM & FIRMS (Collection 61) Giglio et al., 2018.

License: The code in this notebook is licensed under the MIT License. The Environmental Data Science book is licensed under the Creative Commons by Attribution 4.0 license. See further details here.

Contact: If you have any suggestion or report an issue with this notebook, feel free to create an issue or send a direct message to environmental.ds.book@gmail.com.

from datetime import date

print('Notebook repository version: v1.1.0')
print(f'Last tested: {date.today()}')

Notebook repository version: v1.1.0
Last tested: 2025-04-21

References¶

Giglio, L., Schroeder, W., Hall, J. V., & Justice, C. O. (2018). MODIS Collection 6 Active Fire Product User’s Guide Revision B [Techreport]. NASA. https://modis-fire.umd.edu/files/MODIS_C6_Fire_User_Guide_B.pdf
MODIS Science Team. (2017). MOD021KM MODIS/Terra Calibrated Radiances 5-Min L1B Swath 1km. NASA Level 1. 10.5067/MODIS/MOD021KM.061