2018: New activities on atmospheric composition retrieval from satellites – Moving from OMI with GRS-TU Delft & KNMI to Sentinel-3 with EUMETSAT

2017 was an intensive year for my research activity focused on validating the aerosol height retrieval using neural networks (NN), and improving the aerosol correction in tropospheric NO2 determination. Both from the Ozone Monitoring Instrument (OMI) satellite. What was new in 2017?:

  • our paper on the design of the OMI aerosol layer height (ALH) NN algorithm was published here
  • the next paper on its evaluation on specific cases was submitted and is currently under review
  • I am happy to have co-authored the work of Proestakis et al., on a new desert dust aerosol climatology dataset over Asia from 9-year CALIOP observations here
  • I had the opportunity to share the outcome of our activities at 5 conferences and seminars here
  • we attended the successful launch of TROPOMI on-board Sentinel-5 Precursor and the first impressive trace gas and aerosol images here
  • we are preparing a new and last paper on the improved aerosol correction when retrieving tropospheric NO2 from OMI.

I have enjoyed a great collaboration and success with the Geoscience & Remote Sensing (GRS) department, at the Delft University of Technology (TU Delft) and the Royal Netherlands Meteorological Institute (KNMI). But time has come to move forward, and 2018 (and the next years) will be devoted to another satellite mission, in a new country.

Still in science, with a keen eye  on atmospheric composition observations from satellites of course!

I am very glad to join the European Organization for the Exploitation of Meteorological Satellites (EUMETSAT), based in Darmstadt Germany, and to join the scientists and groups working on the development of the atmospheric products (aerosols, fires, H2O, etc…) from the Copernicus Sentinel-3 space-borne mission. If you are interested by this topic, stay tuned and keep following this blog and my twitter account. I will continue sharing exciting findings, news, maps, etc…

Of course, I will still keep a very interested eye on all the other satellite observations (in particular TROPOMI) and groups working on similar topics. I take the opportunity to emphasise how honoured and enthusiastic I am to have worked during 4 years with KNMI and GRS – TU Delft. In particular, I am very grateful to my Promotors Prof. Dr. Pieternel Levelt, Dr. Pepijn Veefkind and Dr. Tim Vlemmix. I do not forget all my colleagues and friends who closely or remotely shared this journey!

Last but not least, 2018 should witness the awarding of my PhD doctoral degree, related to my OMI, aerosol and NO2 research work, from TU Delft and KNMI. Gotten curious about the book to be published in the next months? Here just an illustration of the expected cover (thanks to my friend Remi Charton for his design talents!). Stay tuned!

Expected cover of my (draft) PhD book GRS-TU Delft / KNMI to be published in 2018 (Credits Remi Charton for the design).


A new desert dust aerosol climatology from 9-year CALIOP observations over Asia – Very happy to co-author new ACP paper of Proestakis et al. (2018), from National Observatory of Athens (NOA)!

Spatial distribution of the seasonal mean CALIPSO AOD, D_AOD, optical depth difference between AOD and D_AOD (non-dust AOD), and the percentage of D_AOD with respect to the total AOD for the domain between 65–155 E and 5–55 N and for the period January 2007–December 2015 (Proestakis et al., 2018)

A recent work achieved by my colleagues of the National Observatory of Athens (NOA): 9 years of observations acquired by the CALIOP space-borne instrument were combined with EARLINET ground-based measurements to provide a climatology of desert dust particles over South and East Asia.

Having such a knowledge is important for many research studies focused on atmospheric transport and climate effect of dust.

You can see more details on my webpage here, and the paper of Proestakis et al. (2018) published in the Atmospheric Chemistry and Physics (ACP) journal here.

2017 seen from the sky – A personal selection

2017 is about finishing, 2018 is going to take over. Here is a personal retrospective view on 2017 with a subjective selection of remarkable satellite images.

Wildfires across the globe

Massive wildfires spread across parts of the world such as Chile; Western Canada and the United States; Portugal and Spain; France and even Greenland in 2017. The most recent once, California’s Thomas Fire, was even the largest wildfire of the year in the state. MODIS AquaAqua, Sentinel-2 and many other space-borne instruments reveal these episodes and their intensity. Due to the on-going changes in our climate, it is unfortunately expected that such fires will occur again in 2018, perhaps even stronger.

Red Sun, dust and smoke in Europe

In October 2017, a unique combination of 2 independent events – fires in Portugal & Spain, and a tropical storm – brought dust and smoke all over Europe leading then an unusual red Sun observed in France, UK, Finland and the Netherlands. These events were observed notably by OMI, GOME-2 and MODIS Aqua. More information in No borders for particles! – Red Sun, Saharan dust and smoke over whole Europe

A huge 100-Mile Iceberg breaking away from Antarctica

An iceberg of 120 miles long and with a weight of more than a trillion metric tons broke away in Antarctica last July. It was floating along the so-called Larsen-C ice shell and was carefully monitored by worldwide scientists, notably by Sentinel-1 images. This event rose even more concerns about how Antarctica is evolving in the context of a warmer climate and sea level perturbations.

Such event was well documented in near-real time by our colleague Stef Lhermitte from GRS-TU Delft.

Sentinel-1 B image acquired on 12 July 2017 shown the break off of he Larsen-C ice shelf, spawning one of the largest icebergs on record and changing the outline of the Antarctic Peninsula forever. Credit ESA (Source: http://www.esa.int/spaceinimages/Images/2017/07/Larsen_C_breaks).

California wildfire lights in, Puerto Rico lights out

Satellite images of Earth “night lights” have been a curiosity for the public and a tool of fundamental research for at least 25 years. New global maps have been released by NASA’s Goddard Space Flight Center using the NASA-NOAA Suomi National Polar-orbiting Partnership (Suomi-NPP) satellite. Some of them depict in an impressive way some dramatic episodes of this year: Hurricane Maria struck Puerto Rico, a commonwealth of the United States, leaving the island devastated and all but destroying its power grid; the fast-moving fire that swept into Ventura, California, on December 5, 2017. looking as bright as the urban Los Angeles area.

Monitoring hurricanes

Given the important images, it is important to monitor the mentioned hurricanes and tropical storms. Satellites like Sentinel-3 allow not only to characterise them (e.g. in terms of brightness temperature) but also to follow their displacement.

The Ophelia storm as observed by the Sentinel-3 A satellite. Left: image on 11.102017, when Hurricane Ophelia was about 1300 km south-west of the Azores islands and some 2000 km off the African coast. Right: Brightness temperature of the clouds at the top of the storm, some 12–15 km above the ocean, range from about –50°C near the eye of the storm to about 15°C at the edges on17.10.2017. Copyright ESA (Source: https://dragon3.esa.int/web/guest/missions/esa-eo-missions/sentinel-3/news/-/article/monitoring-hurricane-ophelia).

First air quality images from TROPOMI aboard Sentinel-5 Precursor (P)

Of course, how not to finish this series without mentioning Sentinel-5 P and its instrument TROPOMI? Many persons waited for it for more than 10 years. I personally heard about it the first time some 5-6 years ago, and then waited for its launch during the last 4 years while working on my research with its predecessor OMI. And finally, just a few months before the end of my current job, Sentinel-5 P was launched, a very happy Friday 13th (October)!

With its unique high spatial resolution (7 x 3.5 km2) and top design sensor, this is the very first European operational mission dedicated to atmospheric composition within the Copernicus programme. The instrument is still under calibration until April 2018, being prepared for the developed services, and many works have to be done to derive accurate estimations of toxic gases and particles in the atmosphere. But the very 1st revealed maps of NO2 – Nitrogen dioxide, SO2 – Sulphur dioxide, O3 – Ozone, CO – Carbon monoxide are already outstanding. What is mostly remarkable are the very fine scale structures of all the pollution plumes already visible.

No doubt that the next months and years will be promising for this instrument. Air pollutants will be seen in much more detail than what has been accomplished before. And, while these first results demonstrate the sophistication of the satellite’s instrument, they certainly bring the issue of air pollution sharply into focus. Even though I am going to work on another project in 2018, I will keep a close eye on the next results from TROPOMI S5P.

And in addition look at this NO2 global animation – an average over a couple of days from TROPOMI S5P. Credit KNMI / ESA here.



And so many other pictures of 2017 that I could have posted here, but I invite you to have a look to my past posted weblog pages.

Looking forward to more images and discoveries from space in 2018! And I wish to everyone reading this a Merry Christmas and a very healthy & happy new year!

Must read – A warning from 15,000 world scientists


25 years ago, in 1992, 1700 independent scientists already told us in their “World Scientists – Warning to Humanity – 1st Notice”. And yet, it seems that was not sufficient. This month, a second notice was submitted, this time by about 15,000 worldwide scientists. And the message is still quite clear: “the world is going toward catastrophic biodiversity loss and untold amounts of human misery” and we have little time to act before avoiding this.

This may sound like an unrealistic, even irrational alarmist message. But it is worth reminding that, since its beginning, science work (and scientists) is beforehand based on observations, facts, interpretations with evaluation of uncertainties, implementing models that are then continuously and thoroughly evaluated  by confronting with observations.

This article is relatively short and very simple to read. No need to be an expert, every citizen can easily understand all the messages. I think this is the duty to each of us to take some minutes for reviewing it, and then thinking about our actions. Especially, the causes are nowadays well identified: rapid population growth, deforestation, agricultural production, and rising greenhouse gases from burning fossil fuels that has triggered and driven our climate change.

For this last one, in spite of continuous denial and unjustified claims (especially in the USA), the odds from IPCC and COP23 are pretty clear: CO2 – Carbon dioxide and CH4 – Methane amount in the atmosphere are historically unique, and the emissions keep increasing. Some of these increases are also even observable from space.

Atmospheric levels of CH4 – Methane from 2003 to 2005 and 2008 to 2010, showing  increased concentrations in the latter dataset (in red). The newly released ‘Climate Research Data Package No. 3’ covers more than one decade (2003–14) of atmospheric data products used to get information on the sources and sinks of carbon dioxide and methane. The data products are available through the website: http://www.esa-ghg-cci.org/ (Source ESA: http://www.esa.int/Our_Activities/Observing_the_Earth/Methane_and_carbon_dioxide_on_the_rise).


A positive sign: stratospheric O3 – Ozone hole over Antarctic is recovering. If our actions have allowed to create and then solved this problem, then we can still do something, at every level, about the greenhouse gases.

Several journals and websites forwarded this article. But I have the feeling yet it could reach more and more people and lead to further discussions and actions in our societies. Hope this is not too late…


More information?

  • Ripple et al., 2017: Full articles World Scientists’ Warning to Humanity, 1st and 2nd notices here. You can also endorse the 2017 article here
  • European Space Agency (ESA): CO2 – Carbon dioxide and CH4 – Methane on the rise here
  • CO2 – Carbon dioxide WebPage here
  • CH4 – Methane WebPage Here


No borders for particles! – Red Sun, Saharan dust and smoke over whole Europe

What happened on Monday 16th October 2017? From diverse countries (France, United Kingdom, the Netherlands, Finland), people reported to have observed a “red Sun”. This is a direct result of natural events, although issued from different sources and locations, which combined together over the whole Europe.

During a couple of days, between 16th and 18th October 2017, northern Portugal and Spain were victims of violent wildfires ravaging parts of their territory. These fires released large amounts of fine black particles, also named aerosols, that can absorb the Sun light.

Aerosol optical depth (AOD) (550 nm) simulated by CAMS forecasts of aerosol optical depth (AOD) show enhanced AOD extending from the Iberian peninsula CAMS) on 2017.10.16 (Source: http://atmosphere.copernicus.eu/news-and-media/news/saharan-dust-and-smoke-over-france-and-ukhttp://atmosphere.copernicus.eu/news-and-media/news/saharan-dust-and-smoke-over-france-and-uk).

The Copernicus Atmosphere Monitoring Service (CAMS) forecasts of aerosol optical depth (AOD) show high values extending from the Iberian peninsula to the British Isles.

In addition to these fires, a tropical storm, named Ophelia, appeared 1300 km south-west of the Azores islands and some 2000 km off the African coast. Originally classified as a tropical storm, it was upgraded to a hurricane. The storm moved north-easterly, towards Spain and Britain, collecting sand from the Sahara desert. The related dust particles were then mixed with the black carbon from Portugal fires. Their scattering properties with the solar light led to this red Sun observed by many Europeans, instead of its natural yellow colour in clear sky or white/milky in presence of thin cloud.

The Ophelia storm as observed by the Sentinel-3 A satellite. Left: image on 11.102017, when Hurricane Ophelia was about 1300 km south-west of the Azores islands and some 2000 km off the African coast. Right: Brightness temperature of the clouds at the top of the storm, some 12–15 km above the ocean, range from about –50°C near the eye of the storm to about 15°C at the edges on17.10.2017. Copyright ESA (Source: https://dragon3.esa.int/web/guest/missions/esa-eo-missions/sentinel-3/news/-/article/monitoring-hurricane-ophelia).

The dust particles can be observed through the visible colour composite image from the MODIS instrument, on-board Aqua, on 16.10.2017: some yellow colours are mixed with some thin clouds.

MODIS-Aqua visible image on 16.10.2017 (left) and 17.10.2017 (right) acquired in early afternoon: dust particles in yellow over France, united Kingdom and the Netherlands (Source: https://worldview.earthdata.nasa.gov).

It remains, overall, challenging for satellite measurements acquired in the visible spectrum to easily distinguish dust particles from transparent clouds or cirrus.

Aerosol Absorbing Index (AAI) from GOME-2 sensor on-board MetOp-B platform on 16-17-18.10.2017. Copyright KNMI, EUMETSAT and the Atmospheric Composition (AC) Satellite Applicattion facilities (SAF) (Source: http://www.temis.nl/o3msaf/vaac/vaac_gome2.php?sat=gome2b&year=2017&datatype=pics&freq=daily&filter=Filtered&region=Europe).

GOME-2 and OMI satellite sensors also reveal through the AAI index, with high values in red, the presence of black absorbing particles (i.e. smoke from biomass fires in Portugal) in large quantity. These particles were released in the North of Portugal before being transported to the North (UK) and then East (the Netherlands, Finland and Russia).

Aerosol Absorbing Index (AAI) from OMI sensor on-board the Aura platform on 16-17.10.2017. Copyright KNMI, FMI and NASA. (Source: http://www.temis.nl/airpollution/absaai/absaai-omi.php?year=2017&datatype=pics&freq=daily).

Although the aerosol particles were the visible part of the pollutant transport, IASI sensor revealed the additional presence of CO – Carbon monoxide, a toxic gas issued from incomplete biomass combustion by the fires. CO is a gas pollutant that cannot be visible in the eyes. It can only be measured in the shortwave or thermal infrared spectrum such as the IASI measurement.

Total CO – Carbon monoxide column observed by the IASI sensor on-board MetOp platforms. Copyright LATMOS-ULB & AC SAF (Source: http://acsaf.org/index.html).

This impressive pollutant transport is quite unique due the combination of heterogeneous sources at different locations. But, similar episodes of particle and gas pollution were observed last Summer in Canada as well, as described in our last post Canada wildfires from space and ground – Seeing beyond the flames: a series of observations.

All these worldwide satellite pictures very well illustrate that, although emissions can be national, mix of pollutants and their transport are not contained within the limits of borders. This shows how much pollution and their scientific and societal challenges are an international concern!

More information?

  • CAMS, Saharan dust and smoke over France and UK here
  • GOME-2 AAI index by KNMI, EUMETSAT AC SAF & TEMIS here
  • OMI AAI maps on the TEMIS website here
  • The Ophelia storm as seen by the ESA Sentinel-3 mission here
  • OMI sensor here & IASI sensor here
  • Canada wildfires from space and ground – Seeing beyond the flames: a series of observations WebPage here

Don’t miss #GRS_TROPOMI event, a series of post on “why are we excited this week about #TROPOMI #Sentinel5P launch?!”


Why are we (I and my colleagues) that excited this week? Why is TROPOMI so much important for the new coming era in air quality & climate satellite era?

Throughout the week, we will post more information on our GRS TU Delft website and also be using the #GRS_TROPOMI on social media (Twitter and Instagram) as we explain more about the mission, its goals, and how it all works in relation to the goals and work being done in our department. The week will culminate with the Sentinel-5P launch event taking place at the Space Expo in Noordwijk.


More information?


20th OMI Science Meeting at NASA Goddard Space Institute

During 3 days, from 12 to 14 September 2017, I had the pleasure to attend, with my KNMI colleagues from the R&D satellite department, the 20th OMI Science Team Meeting hosted by NASA Goddard Space Institute in Greenbelt MD (close to Washington DC) in the USA.

Group picture at NASA, Goddard Space Institute, Greenbelt MD, Virginia, USA, 2017.09.13 in front the morning and A-Train satellite constellation, with all our OMI colleagues.

Despite its quite advance age for a satellite mission (13 years old!), OMI is still delivering remarkable measurements about our atmospheric composition and air quality. So many talks and discussions on the aerosol global record over cloud-free scenes and above clouds, decade global volcanic SO2 – Sulfur dioxide missions, the use of OMI data by air quality model simulations to inform air quality policy, the case studies on emissions monitoring and to support authorities and clean-tech industry, the new generation of the Quality Assurance For Essential Climate Variables (QA4ECV), the evolution in the ozone trends and related mechanisms, and of course the future with the forthcoming TROPOMI (Sentinel-5 Precursor) mission, TEMPO (NASA Geostationary) and TROPOLITE.

I was honoured to present the status of our work on the retrieval of aerosol layer height from the OMI visible band using machine learning technique, and the results with single day CALIOP aerosol along-track observations.

Chimot J., Veefkind P., van Ouwerkerk G., Vlemmix T., Levelt P., Aerosol layer height retrieval from OMI and neural network- Possibility for a 13-year time series?, 20th OMI Science Team Meeting, NASA Goddard, Greenbelt MD, Virginia, USA, 2017.09.12 (Source: https://www.researchgate.net/publication/319678283_Aerosol_layer_height_from_OMI_and_neural_network_Possibility_of_a_13-year_time_series)

Several nice social moments accompanied us during these 3 days:

  • The Nationals baseball game on Wednesday 13th September evening in Washington DC
  • The guided tour by NASA during lunch break, a great opportunity to see the control rooms of the Aura (OMI spacecraft!!), Aqua and Landsat missions!


I am very grateful to all the organizers for this inspiring meeting, and my current promotors (Dr. Pepijn Veefkind, Prof. Dr. Pieternel Levelt & Dr. Tim Vlemmix)!

In spite of being glad of having been part of this adventure, I cannot stop myself thinking this may have been my very last OMI conference, before finishing my current research project and starting new professional & personal adventures (still in satellite & atmospheric community of course!). But this last point will be specifically mentioned later in future weeks. Stay tuned!


More information?