IWV comparison

IWV comparison#

ORCESTRA focuses on the area of the Inter-Tropical Convergence Zone (ITCZ), which we’ve commonly identified by areas of high integrated water vapour (IWV). We’ve used this quantity extensively during Flight Planning and several instruments measuring IWV have been deployed. This notebook compares a few instruments which measured IWV during the campaign.

We’ll first import some required libraries and set up the root path to the ORCESTRA data collection to a stable version. If you are instead interested in the most recent updates, just use the ipns link instead.

import xarray as xr
import matplotlib.pylab as plt

# root = "ipns://latest.orcestra-campaign.org"
root = "ipfs://QmU8L1tadPd7E7h8Rz3fmoF9DYYWj33NuBRqsEwGqCeWBp"

relative occurrence over entire campaign period#

We’ll compare data from the GNSS sensors on RV Meteor as well as the microwave radiometer (HAMP) and the dropsondes from HALO. The three instruments all employ different sensing methods and sampled different areas of the observation region. Still, they overall sampled a similar regime and thus on average, we might expect similar results. This can be confirmed by the following histogram:

sources = [("METEOR GNSS", "METEOR/GNSS_IWV.zarr", "iwv"),
           ("HALO HAMP", "HALO/iwv/*.zarr", "IWV"),
           ("HALO dropsondes", "HALO/dropsondes/Level_3/PERCUSION_Level_3.zarr", "iwv")]
for label, path, var in sources:
    ds = xr.open_mfdataset(f"{root}/products/{path}", engine="zarr")
    plt.hist(ds[var], bins=40, range=(0,100), density=True, histtype="step", label=label);
plt.xlabel("IWV / kg m-2")
plt.ylabel("relative occurance")
plt.legend();
_images/8dec9124cea8c76de7cf9b3ed7dc309ac7ed5964b827133099551bea6b7c7824.png

Note

At least some of the data is still in early processing stages, so you still might find artifacts which should be removed in subsequent stages of quality control.

Although the instruments broadly agree, we can already see some features of the measurement principle. E.g.:

  • The microwave retrieval (HAMP) produces artificially high values when flying over land (instead of ocean). These values have not yet been removed and are visible in the histogram.

  • Sometimes, the dropsondes didn’t measure humidity at all vertical levels (sometimes even none at all). These values have not yet been masked in the current dataset.

  • The lower IWV values of hamp (e.g. < 30 kg/m2) correspond to the 3rd phase of ORCESTRA, which was based in Europe (see next figure). This period was not covered by RV Meteor and used fewer dropsondes.

import cartopy.crs as ccrs
import cartopy.feature as cfeature

ds = xr.open_mfdataset(f"{root}/products/HALO/iwv/*.zarr", engine="zarr")   
low_iwv = ds.where(ds.IWV.compute() < 30, drop="True")

fig = plt.figure()
ax = fig.add_subplot(1, 1, 1, projection=ccrs.PlateCarree())
ax.add_feature(cfeature.COASTLINE)
plt.plot(low_iwv.lon, low_iwv.lat, ".", ms=1, transform=ccrs.PlateCarree())
plt.title("HALO HAMP, iwv<30");
_images/d43cf8b5fcb13d6b2aa07b71913586d0d419c3262f2c3416ea28a4d1ac91dc0b.png

data sources#

The datasets have been provided by various individuals and institutions. Please inspect the dataset attributes and contact the authors in case you use the data for scientific work.

Hide code cell source 
for label, path, var in sources:
    ds = xr.open_mfdataset(f"{root}/products/{path}", engine="zarr")
    print(label)
    for k, v in ds.attrs.items():
        print(f"{k}: {v}")
    print()

METEOR GNSS
contact_person: Pierre Bosser (pierre.bosser@ensta-bretagne.fr)
description: IWV estimates for R/V Sonne
history: Created on 20241216_120643; Converted to zarr by Lukas Kluft (lukas.kluft@mpimet.mpg.de
institution: ENSTA Bretagne
methodology: ZTD are screened using range check and outlier check on both position and ZTD. ZTD are converted to IWV using ERA5 sea level pressure data, 1-hourly, 0.25x0.25 deg and VMF1 atmospheric integrated mean temperature, 6-hourly,  2.0x2.5 deg. See [Bosser et al., 2021, ESSD] for more information.
processing_software: GIPSYX v2.2 in PPP mode
resolution: IWV temporal resolution = 30 s
title: IWV data from GNSS antenna on R/V Meteoor
version: v1.0

HALO HAMP
Comment:  
Host-PC_Software_Version: V9.45
Radiometer_Location: HALO KV
Radiometer_Software_Version: V9.45
Radiometer_System:  RPG-HATPRO
Serial_Number:  R-DPR-19/007
Station_Altitude:  200
Station_Latitude: 48 05.05
Station_Longitude: 11 16.75
georeference source: BAHAMAS IGI 100 Hz
netCDF_Convention:  CF-1.0
version: 0.3

HALO dropsondes
authors: Helene Gloeckner, Theresa Mieslinger, Nina Robbins
broken_sondes: ['233814577', '234141211', '233814605', '233814531', '233441164', '233814590', '233814578', '234141210', '233814536', '233814535', '233814584', '233814586', '234141208', '233814546', '234141209', '233814537', '233814533', '140238', '140529', '140425', '180450', '231220385', '234021411', '180732', '233530211', '234030059', '180136', '231820683']
contact_email: helene.gloeckner@mpimet.mpg.de
featureType: trajectoryProfile
history: ASPEN processing with Aspen V4.0.2 
quality control with pydropsonde 0.0.0.post406.dev0+e15221f 
Level 3 processing with pydropsonde 0.0.0.post406.dev0+e15221f 
2024-12-04T11:54:54.877787+00:00 level3 concatenation with pydropsonde 0.0.0.post406.dev0+e15221f 

institution: PERCUSION subcampaign of ORCESTRA 2024
references: https://orcestra-campaign.org/orcestra.html,https://github.com/atmdrops/pydropsonde
source: dropsonde and postprocessing using pydropsonde
title: Dropsonde data from the ORCESTRA field campaign