Usecase_6_CO_European_wildfires

Demonstrating new plotting functions in Atmospheric Virtual Lab¶

This notebook demonstrates the new plotting functionalities in the atmospheric-virtual-lab (avl) module. The atmospheric-virtual-labis a toolkit developed for interactive vislualization of atmospheric satellite data.

Here we demonstrate the MapPlot and MapPlot3D commands, and create interactive maps of Copernicus Sentinel 5p / TROPOMI carbon monoxide (CO) observations from European wildfires on 18th July 2022. These plotting commands create a Jupyter widget called ipyleaflet which a very convenient way to instantly visualize HARP products, including single orbits.

1. Create new conda environment for avl ¶

The easiest way to get started with atmospheric-virtual-lab is to create a new conda environment and pre-install the required packages. Here we create a new conda environment called avl_env, and install required packages as follows:

$ conda create -c conda-forge -n avl_env atmosphere-virtual-lab

Before running the jupyter lab and this notebook the environment needs to be activated:

$ conda activate avl_env

And after that the Jupyter lab can be launched:

$ jupyter-lab

2. Import S5p TROPOMI CO data with harp ¶

First we need to import the needed packages, and define the CO file that we want to plot. The TROPOMI CO file can be downloaded from Sentinel-5P Pre-Operations Data Hub. We are using the sentinelsat python package here to perform the download for us.

# atmosphere-virtual-lab is imported with the name avl
import avl 
import harp
import eofetch

filename_tropomi = 'S5P_RPRO_L2__CO_____20220718T123656_20220718T141826_24674_03_020400_20230113T012004.nc'
eofetch.download(filename_tropomi)

Next the operations for importing the TROPOMI CO product are defined, and the data is imported with harp. In this example we will use the carbonmonoxide_total_column_corrected variable instead of the default carbonmonoxide_total_column. From TROPOMI CO processor version 2.x.x onwards the fixed masked destriping method has been implemented to correct for the "stripes" visible in the CO observations (probably due to calibration issues), and a new parameter carbonmonoxide_total_column_corrected has been included. To use this new parameter, in harp import a new option of co=corrected needs to be added. The recommended quality assurance value for CO is qa>0.5 (with harp 50), but it is also recommended to take a look of the different qa value definitions explained in the product Readme file.

operations = ";".join([
    "CO_column_number_density_validity>50",
    "derive(CO_column_volume_mixing_ratio {time} [ppbv])",
    "keep(latitude_bounds,longitude_bounds,CO_column_number_density,CO_column_volume_mixing_ratio)"
    ])
    
tropomi_co=harp.import_product(filename_tropomi, operations, options="co=corrected")

print(tropomi_co)

source product = 'S5P_RPRO_L2__CO_____20220718T123656_20220718T141826_24674_03_020400_20230113T012004.nc'
history = "2024-05-17T14:37:52Z [harp-1.21] harp.import_product('S5P_RPRO_L2__CO_____20220718T123656_20220718T141826_24674_03_020400_20230113T012004.nc',operations='CO_column_number_density_validity>50;derive(CO_column_volume_mixing_ratio {time} [ppbv]);keep(latitude_bounds,longitude_bounds,CO_column_number_density,CO_column_volume_mixing_ratio)',options='co=corrected')"

float latitude_bounds {time=351103, 4} [degree_north]
float longitude_bounds {time=351103, 4} [degree_east]
float CO_column_number_density {time=351103} [mol/m^2]
double CO_column_volume_mixing_ratio {time=351103} [ppbv]

3. Plotting 2D map with avl ¶

First the imported harp object will be plotted on a 2D interactive map. For this we use avl.Geo command. Within the command we can define the colormap (e.g. default matplotlib color maps), min and max values of the color range as well as the opacity of the overlayed TROPOMI data. Once the map is plotted, the colorscale can be also adjusted manually, by clicking on the palette icon at the lower right corner of the map. In this example we will plot the CO data as ppbv, i.e. use the variable CO_column_volume_mixing_ratio.

variable='CO_column_volume_mixing_ratio'
avl.Geo(tropomi_co, variable,colormap="rainbow",colorrange=(0, 200), opacity=0.7)

Now you can zoom in the interactive map to see the CO emission plumes from the wildfires in France, Spain, and Portugal. The data is automatically overlayed on top of the OpenStreetMap so that locating the plumes geographically e.g. close to cities is easy. The zoom level as well as the center latitude and longitude can be also set in the plotting command:

avl.Geo(tropomi_co, variable,colormap="rainbow",colorrange=(0, 200), opacity=0.7, centerlat=44, centerlon=-10, zoom=5)

4. Plotting 3D map with avl ¶

Next we will use the avl.Geo3D command to plot the TROPOMI CO data onto an interactive 3D globe basemap. The avl.Geo3D command is very similar to avl.Geo:

avl.Geo3D(tropomi_co,variable,colorrange=(0, 200),colormap="rainbow", opacity=0.7,showcolorbar=True)