Working from home, Year 2!

Here we go for a 2nd year working from home again with EUMETSAT and Copernicus Sentinel-3 !

Although we have some hopes for a better health situation during 2021, it is clear we will have to continue to face turbulent moments.

I wish you All a very happy new year. Good luck to everyone going back to work! Let’s hope for a smooth transition from virtual meetings towards more and more real human interaction!

More information on:

  • The European Meteorological Satellite Agency EUMETSAT here
  • The Sentinel-3 mission here
  • The European Copernicus program here

Clouds with Copernicus Sentinel-3 SLSTR over Amazonia

RGB picture from Copernicus Sentinel-3 SLSTR-A over Amazonia on 30.11.2020. SWIR (1.6 um) in Red, and Visible (675 & 550 nm) in Blue & Green. Credit Picture from the SNAP tool box ESA/Brockmann Consult.

There is not only one type of clouds, but many of them, with diverse properties leading to various ways to visualize them when looking at our satellite measurements. Here an example with the Copernicus Sentinel-3 (S3) A Sea & Land Surface Temperature Radiometer (SLSTR) over Amazonia.

The Red-Green-Blue (RGB) picture below, of 30.11.2020 over Amazonia, combines the ShortWave InfraRed (SWIR) 1.6 um and the visible (674 & 550 nm). Water clouds with small droplets largely reflect at all channels & appear white. Snow & ice clouds strongly absorbs SWIR & appear cyan.

What about their altitude?

SLSTR has a special channel at 1.3 um in which water vapor (H2O) so much absorbs that no signal at the top of the atmosphere may be measured. Unless clouds at a very high altitude are present, shielding then a large fraction of the H2O column in the atmosphere. Consequently, a bright signal spikes up and elevated clouds shine!

Radiance at 1.3 um from Copenicus Sentinel-3 SLSTR-A over Amazonia on 30.11.2020. Credit Picture from the SNAP tool box ESA/Brockmann Consult.

From the picture above, snow & ice clouds are primary the highest one! Anything else lower, such as surface and fractional low clouds, seems “invisible”. They cannot be “seen”.

Save the date – Next S3VT on 10-12 March 2020 at EUMETSAT, Germany

Save the date! The next Sentinel-3 Validation Team (S3VT) workshop will be held at EUMETSAT, in Darmstadt, Germany, from the 10th to the 12th of March, 2020! This is an important meeting, jointly chaired by EUMETSAT and ESA, for all scientists & users engaged with validation and evolution recommendations of current & upcoming products from the Copernicus Sentinel-3 mission. And it will be my pleasure to co-chair our 2nd S3VT Atmosphere session.

Anyone who wishes to be involved in the Sentinel-3 validation activities with a privileged access to early new information is very encouraged to submit a proposal to join the S3VT team here. Looking forward to fruitful exchanges on aerosols, fire, H2O, and clouds!

D6MMyBJU8AAW0Hw

 

More information

  • The ESA EUMETSAT S3VT WebPage here
  • Information about the S3VT Atmosphere sub-group here
  • To submit a Valid project idea proposal and join the S3VT as new member here
  • The EUropean organization for the expoitation of METorological SATellites here
  • The European Space Agency ESA here
  • The Copernicus programme from the European Union (EU) here

Winter has come in Europe – UV dose estimated from space

No doubt winter has come in Europe! The Earth inclination has changed.

As observed last Winter with GOME-2 UltraViolet (UV) measurements, a big change in the effective UV irradiance reaching the Earth surface is also seen over the day between last July and nowadays.

Here below is an example with the estimated reduced Vitamine D production in our skin as a direct consequence. These pictures are from the ESA / KNMI Tropospheric Emission Monitoring Internet Service (TEMIS). Satellite UV dose is computed from the assimilated global O3 – ozone field at local solar noon, and with surface downwelling solar (SDS) radiation & cloud information measured by the Meteosat Second Generation (MSG) satellites led by the EUMETSAT agency.

 

More information

  • WebBlog on Vitamin D estimated from GOME-2 UV measurements here
  • Clear-sky UV index forecast from the Tropospheric Emission Monitoring Internet Service (TEMIS) here, Credit ESA / KNMI

How many truckloads for carrying a typical “boomerang” Saharan dust over the Atlantic?

Saharan dust outbreak often occurs across the Atlantic. Here, a big #aerosol plume was captured by  the Sentinel-3 Ocean & Land Colour Instrument (OLCI) from the  Copernicus programme led by the European Union (EU). A big fraction of it usually travels across-Atlantic and one may think it doesn’t come back to us…

Screenshot 2019-11-01 at 22.17.49.png
Red-Green-Blue (RGB) composite imaged from Sentinel-3 OLCI, 2019.11.01. Source: https://s3view.oceandatalab.com/?date=1572609600975&timespan=1d&zoom=3&extent=-10879740.856484%2C-7054220.4654004%2C10879740.856484%2C7054220.4654004&products=3857_Sentinel-3_OLCI_true_RGB%2C3857_Sentinel-3B_OLCI_true_RGB&opacity=100%2C100&stackLevel=85%2C85.01

But the forecast of Aerosol Optical Depth (AOD) for Friday 1st November 2019, computed by the Copernicus Atmospheric Service (CAMS) & depicted by the Windy on-line application, shows a striking dust “boomerang”-like pattern. After travelling a large distance, a heavy dust load eventually comes back to South Europe. Wind is a key factor explaining such aerosol long-range transport.

Screenshot 2019-11-01 at 22.23.47.png
AOD forecast for Friday 2019.11.01 at 11 am, computed by CAMS and illustrated by Windy. Source: https://www.windy.com/-Menu/tools?aod550,15.708,-55.020,3

My colleague, Dr. Antti Lipponen, researcher at the Finnish Meteorological Institue (FMI), evaluated this transport phenomenon in a fancy unit. “Based on NASA’s GEOS-FP analysis the area of the dust over the Atlantic Ocean is about 9.1 million km² and the mass of the dust about 2.6 million tons (2 600 000 000 kg). If a truck would carry 40 tons of dust it would make about 65 000 truckloads!!!

Screenshot 2019-11-03 at 14.50.10.png
Source: https://twitter.com/anttilip/status/1190603602819256321

More information?

  • Windy here
  • Copernicus Atmospheric Service (CAMS) here
  • Follow Anti Lipponen on Twitter here
  • Aerosol WebPage here

High fire risks in California

Fire risks are quite high these days in Southern California. And some fires have even already been observed  by the Sentinel-3 (S3) B satellite, from the European Copernicus programme. Here below are some Red-Green-Blue (RGB) pictures from the Level 1 (L1) measurements of the optical S3B sensors, disseminated by EUMETSAT: the Ocean and Land Colour Instrument (OLCI) and the oblique view of the Sea & Land Surface Temperature Radiometer (SLSTR). These images show smoke  aerosol particles spreading from North of Los Angeles over Pacific.

As illustrated by @weatherchannel, these areas are located in “Critical” fire danger zones.

Screenshot 2019-10-13 at 16.18.02.png

 

More information:

  • Aerosol WebPage here
  • The European Union (EU) Copernicus programme here
  • The European Meteorological & climate satellite (EUMETSAT )agency here
  • The Copernicus Sentinel-3 mission operated by EUMETSAT here

Saharan dust transport over Gran Canaria islands & Cabo Verde

On 2019.02.05, a remarkable dust outbreak issued from the Western Sahara coast spread over Gran Canaria islands. This thick plume, with heavy load of particles, and larger than 1.000 km width, was well observed via a series of satellite images:

 

 

Screenshot 2019-04-21 at 10.39.08
Tweet from the AC SAF showing images from Meteosat Second Generation (MSG) SEVIRI (left) and GOME2 Metop UVAI index (right) of the Saharan dust outbreak of 2019.02.05. Source: https://twitter.com/Atmospheric_SAF/status/1092847004844208128

Several days later, on 2019.03.02, another Saharan dust was transported over Cabo Verde. Similarly, a very large and thick plume was captured in the images from the NASA SUOMI VIIRS sensor, and measured by the aerosol index UVAI from Tropomi.

Such aerosol dust events regularly occur in these areas. I always find the related satellite images not only impressive but also informative regarding their spatial scales, intensity and transport.

 

More information?

  • The EU Copernicus programme here
  • Overview of the NASA & NOAA SUOMI mission here
  • The EUMETSAT AC SAF here
  • EUMETSAT agency here
  • TROPOMI WebPage here
  • Current Earth observation satellites with Sentinel-3 & Sentinel-5 Precursor WebPage here
  • Aerosol webPage here

 

Biomass burning in Central Africa: NOx, CO & aerosol animations from Sentinel-5 P & SUOMI observations

Biomass burning is a major source of trace gases & aerosol particles on a regional and a global scale (Seiler and Crutzen, 1980; Logan et al., 1981; Crutzen and Andreae, 1990; Andreae, 1991). Interannual variations in biomass burning within specific regions of the world can be dramatic, depending on factors such as rainfall and political incentives to clear land. The forest fires in Indonesia during 1997–1998 and those in Mexico during 1998, both related to the El Nino Southern Oscillation (ENSO) induced drought, are well known examples of extreme fire events (e.g. Levine, 1999; Nakajima et al., 1999; Peppler et al., 2000; Cheng and Lin, 2001).

The principal biomass burning areas can be observed in the Amazonian region and in central Africa. Among the trace gases released, NO2 – nitrogen dioxide & CO – carbon monoxide abundances can be very high. Satellite observations are a helpful tool for the identification of these sources in the troposphere and to follow their transport. In addition, these intensive biomass burning episodes release a large quantity of aerosol particles, at fine size and with absorbing properties.

Below are the animations of NO2 and CO columns as observed by the TROPOMI sensor, on-board the Sentinel-5 Precursor mission from the European Copernicus program. These animations cover ~1 month of biomass burning over Central Africa. They are extracted from the SentinelHub Earth Observation (EO) browser.

Additionally, you can visualise here animations based on the NASA SUOMI VIIRS observations showing the fire detected pixels (in red) and the detection of fine absorbing particles in large concentrations. Note that SUOMI and Sentinel-5 P are flying together on the same orbit / same track with only a few minutes apart.

 

More information?

  • TROPOMI, on-board the Copernicus Sentinel-5 Precursor satellite, here
  • NO2 – nitrogen dioxide here
  • CO – carbon monoxide here
  • Aerosol particles here
  • Trace gases in the atmospheric composition here

Suffocating air in China during wintertime – How does it look like from space?

In winter time, cold temperature leads to an increase in using heaters of course. And when the electricity source is notably based on coal power plants, then gas emissions (and particles) increase as well leading to higher pollutant concentration.

Air pollution in China is well known. Satellite observations as evidence show a strong increasing trend of NO2 column concentrations since 1995 in China (Irie et al., 2005; Richter et al., 2005; van der A et al., 2006). The main anthropogenic emissions of NOx in China are from transport and coal-fired power plants (Liu et al., 2015; Saikawa et al., 2017; Li et al., 2017). Because of the rapid implementation of new technologies and air quality control regulations for power plants and vehicles in China, their emission factors and activities are also changing with time. In spite of major reductions the last 7 years thanks to very supportive governmental decisions, it still remains an issue.

Pictures below, from NASA MODIS Aqua let us imagine how suffocating this air may remain these days. TROPOMI Sentinel-5 Precursor observations indeed show high concentrations of not only NO2 – nitrogen dioxide, and CO – carbon monoxide. Efforts in reducing emissions due to the electricity generation and vehicles must continue to ensure a better health for the whole population.

Publication in AMT journal of “Minimizing aerosol effects on the OMI tropospheric NO2 retrieval – An improved use of the 477 nm O2-O2 band and an estimation of the aerosol correction uncertainty”

Our last research paper focused on the challenging topics of aerosol layer height (ALH) retrieval from hyperspectral visible measurements, aerosol correction, and tropospheric NO2 retrieval from satellite sensors like OMI was published in the Atmospheric Measurement Techniques (AMT) journal. This work relies on the activities achieved during the last months of my thesis research with my colleagues of the Geoscience and Remote Sensing (GRS) department of TU Delft and KNMI: Dr. J. Pepijn Veefkind, Dr. Johan de Haan, Dr. Piet Stammess, and Prof. Dr. Pieternel  F. Levelt.

This paper is based on the last developments we published during 2016, 2017, and 2018. During these years, not only the OMI cloud algorithm was improved (Veefkind et al., 2016), but also an OMI aerosol layer height (and optical thickness) neural network algorithm was developed (Chimot et al., 2017, 2018). This time, we directly evaluate the impacts of these developments to correct of aerosol absorption and scattering effects in the visible spectral range in view of retrieving tropospheric NO2, an important trace gas affecting air quality in urban and industrialised areas.

What are the main conclusions? Aerosol correction on tropospheric NO2 retrieval from OMI has been greatly improving the last 2-3 years for UV-Vis air quality satellites. Notably thanks to the updated effective cloud retrievals. But also there is a clear potential from the ALH based on the 477 nm O2-O2 band. However, the decision for the future processors is not necessarily easy to take: accuracy vs. radiance closure budget are clearly competing.

Gotten curious? See more information here.

I greatly thank my co-authors from the Netherlands for this very interesting work! This paper closes the loop of my whole research work achieved during the last 4 years with the Geoscience and Remote Sensing (GRS) department of TU Delft and KNMI.

f04
Average maps of MODIS AOD(550nm), OMI DOMINO NNvO2 and differences after applying the implicit (with OMCLDO2-New) or explicit (with NNMODIS,SSA=0.95) aerosol correction over China in summertime (June–July–August) 2006–2007. (a) MODIS AOD(550nm), (b) OMI DOMINO NNvO2, (c) OMI NNvO2 differences due to changes between OMCLDO2-New and DOMINO implicit aerosol corrections, (d) NNvO2 differences between explicit aerosol correction based on the NNMODIS,SSA=0.95 aerosol parameters (i.e. aerosol forward model assuming SSA = 0.95, MODIS AOD(550 nm) and retrieved ALH) and implicit aerosol correction implemented in DOMINO.