THOR - What Happens Above Thunderstorms?
  1. 2015 • ISS Increments 43-44
  2. 2015 • ISS 44S (Soyuz TMA-18M) "IrISS" - short-duration mission
Physical Sciences:
  • Earth observation
T. Neubert (1), N. Larsen (2), O. Chanrion (1), E. Blanc (3), Y. Yair (4)
National Space Institute
Technical University of Denmark (DTU Space)
Elektrovej 328
Kongens Lyngby
Interdisciplinary Center Herzliya (IDC)
NOTE: Andreas Mogensen became Denmark’s first astronaut when he launched on board Soyuz TMA-18M (ISS 44S) on 2 September 2015 for his 10-day iriss mission. He landed with Soyuz TMA-16M on 12 September 2015.

Water vapour transport to the stratosphere powered by thunderstorm convection

Water is an important component of the Earth´s atmosphere and a key element of the climate system:
(1) water vapour in the troposphere and stratosphere acts as a greenhouse gas responsible for a large part of the absorption of outgoing long wave radiation in the atmosphere.
(2) when water condenses into clouds or freezes to ice on the earth’s surface, it modifies the Earth’s albedo, affecting the reflection of energy from the sun back into space.
Water vapour in the stratosphere is an efficient greenhouse gas and may reside for months in this layer with decadal variations which still need to be represented in climate models.
Evidence suggests that these variations are mirrored in the surface winds and temperatures which vary on seasonal to decadal timescales [Shaw, 2008]. Recent research has shown that the sudden drop in stratospheric water content in 2001 reduced the trend in global surface temperature by 25 % in the first decade of the 21th century [Solomon et al., 2010].
The results of these studies underline the necessity of correct representation of the processes controlling stratospheric water content in climate models.

The stratospheric water originates through three processes:
(1) water vapour enters the lowest stratosphere through the tropical tropopause along with slowly ascending air masses. This pathway is thought to be responsible for the major part of water input to the stratosphere. The water content of the ascending air is essentially determined by the temperature and pressure of the last cloud on its path, where the partial pressure of water
vapour was adjusted to the equilibrium water vapour pressure over ice at the cloud temperature. This process is referred to as “dehydration”.

(2) Ice particles enter the stratosphere through deep convective overshoots during tropical thunderstorms.
Occasionally convective up-draft in cumulus nimbus clouds gets so severe that the cloud penetrates the tropopause for a shorter range of time, typically less than an hour, creating the “cloud turrets”. The amount of water vapour released during such an event depends on the amount of ice transported to the stratosphere, and its ability to mix with the surrounding air before the cloud sinks back to the troposphere.

(3) Water is produced in the stratosphere from oxidation of methane. This process is progressing gradually as the air rises through the stratosphere, causing an increase of water vapour content with altitude. This water flux is determined by the methane mixing ratio of the upper troposphere, and is as such well understood.
It is currently being debated to what extent each of these components controls the water input to the stratosphere. Especially the relative importance of the two first components is a subject of discussion. However recent publications suggest that clouds extend further than expected from visual identification and that this aspect is poorly researched, in particular at the top of very high clouds (Koren et al., 2007; Lane and Sharman, 2014).

Circulation in the stratosphere and mesosphere driven by internal gravity waves generated by thunderstorms

Gravity waves are generated in the troposphere from orographic features of the Earth’s surface as the wind is bent when blowing across the features. An example is the wind coming from the Atlantic Ocean meeting the Pyrenees Mountains or the Alps. They can also be generated by thunderstorms from the powerful convection in the thunderstorm core. The waves can travel to the upper troposphere, modulating the cloud cover in wavelike
patterns, and may reach the stratosphere and mesosphere. In the mesosphere they can be detected as wave-like patterns in the air glow layer.

Atmospheric gravity waves exist by virtue of the stable density stratification of the atmosphere under gravity. Disturbances to a balanced state can result in excitation of atmospheric gravity waves with a variety of spatial and temporal scales. Horizontal wavelengths range from kilometres to thousands of kilometres, and periods range from ten minutes in the troposphere to the inertial period, which is infinite at the Equator and 12 hours at the poles. Gravity waves can occur at all altitudes in the atmosphere. They can transport energy and momentum from one region of the atmosphere to another; they can initiate and modulate convection and subsequent hydrological processes (e.g., Mapes [1993]); they disturb the smooth, balanced state through injection of energy and momentum into the flow.

Waves have a large influence on the Mesosphere-Lower Thermosphere region because of their ability to transport significant energy and momentum and to influence larger scale motions. The mean momentum fluxes imply a strong residual circulation near the mesosphere, with upwelling at high summer latitudes and descent in the winter hemisphere which result in temperatures ~90 K above or below radiative equilibrium values and a reversal of the latitudinal gradient of temperature relative to lower altitudes [e.g. Luo et al., 1995]. Although our understanding of gravity waves in the mesosphere has improved considerably in the past decade, quantitative characterization of their role is still in its infancy.

Studies of Transient Luminous Events (TLEs)
The ASIM mission was de-scoped by excluding limb-viewing optical instruments. Astronaut observations from the ISS will allow us to regain some observations. The scientific rationale is discussed in [AD1]. The scientific questions that can be addressed with limb-viewing from the ISS concerns the vertical structure of TLEs. We are in particular interested in jets and gigantic jets, which are lightning reaching upwards through the stratosphere and mesosphere. These have been poorly characterized to date because of poor viewing conditions from the ground, and because they appear to be rarer than the sprites and the elves.
Astronaut observations from the ISS will allow us to regain some observations.

The scientific investigations concern two topics powered by thunderstorm convection which relate to water vapour transport:
1. Transport of water from the troposphere to the stratosphere by thunderstorm convection.
2. Circulation of the stratosphere and mesosphere driven by internal gravity waves generated by thunderstorms. The project will study the relationship between these processes and electric activity of thunderstorms by using optical cameras on the International Space Station (ISS), ground observations of lightning, and meteorological satellite observations of cloud properties.
A third topic relates directly to lightning activity. It concerns one of the main topics of investigation for the ASIM project to the ISS external Columbus platform.
3. Generation of electric discharges in the stratosphere and mesosphere over thunderstorms.

Experiment Specific Goals and Detailed Objectives

Lightning activity is generated by thunderstorm convection and increases with increasing convection. Likewise, the formation of cloud turrets and generation of gravity waves increases with increasing convection. Monitoring of the global lightning activity is a simple, inexpensive and is already done by several ground systems. If one can identify a reliable correlation between lightning activity on one hand and cloud turret and gravity wave formation on the other hand, global observations of lightning activity may be used as a proxy for quantifying water transport to the upper layers of the atmosphere. It will allow their implementation in climate models, increasing their accuracy. Likewise, measurements of TLEs depend on cloud convection. Little is known about their relation to lightning activity over, for instance, the most prolific thunderstorm regions of Africa where space observations of the past have not had the resolution to properly image TLEs and ground observations have not been attempted because of logistical challenges. Likewise, the generation of jets and gigantic jets by thunderstorm fields are not well known.

RO1: understand how lightning activity powers cloud turrets
RO2: understand how lightning activity powers gravity waves
RO3: understand the structure of TLEs above thunderstorms
By: measuring cloud turrets, TLEs, and gravity waves and in the stratosphere and mesosphere above thunderstorms from the ISS, and relating the observations to lightning activity measured by global lighting detection systems and to meteorological satellite observations of cloud properties.
TO1: identify camera system elements and functionalities that optimize observations of cloud turrets, TLEs and gravity waves
TO2: test the experimental concept
TO3: make recommendations for future comprehensive experiments by ISS astronauts
Justification for Need of Space Experiment
There are two main reasons why a space experiment from the ISS is needed. The first relates to the logistics of observations above thunderstorms. Some of the very intense thunderstorm regions are in the tropical and subtropical regions that are difficult to access and therefore pose significant logistical challenges. Here, on the other hand the ISS offers an almost complete coverage with its orbit inclination of 51.6o. In addition, the ISS is the platform that has the lowest orbit available and therefore brings us as close as possible to the phenomena we want to observe.
The second reason is that we will attempt to observe in optical bands that are subject to absorption in the atmosphere. These bands are not available for ground observations.
The phenomena we want to observe occur in the high-altitude atmosphere and can therefore be observed from space in the proposed bands. In the case of cloud turrets, the altitude is above 10 km, for TLEs it is 10-80 km and in the case of gravity waves the altitude is above 80 km. The bands adopted are expected to give us new physics.
Gravity waves
• The Atmospheric Emissions Photometric Imager (AEPI ) experiment on the ATLAS-1 (Atmospheric Laboratory for Applications and Science ) space shuttle mission, launched March 24, 1992 (Mende et al., 1994; 1998).
• The High-Resolution Doppler Imager (HRDI) on the UARS (Upper Atmosphere Research Satellite) (Hays et al., 2003).
• The experiment LSO (Lightning and Sprite Observations) during the flight of the French Astronaut Claudie Haignere (mission Andromède) in October 2001 (Blanc et al., 2004; Belyaev et al., 2006) and ESA taxi flights: ODISSEA, CERVANTES and DELTA.

Cloud Turrets
• Aircraft observations over the years.
• Satellite observations with marginal resolution (Nielsen et al., 2011).

• Ground and aircraft observations over the years.
• Satellite observations with poor spatial resolution (ISUAL on FORMOSAT-2).
• GLIMS on the ISS 2012-
It is proposed for an astronaut to take optical images of the earth’s atmosphere with cameras through the windows of the ISS. Observations are taken over thunderstorms.

Cloud turrets
We are primarily interested in the top of the clouds (8-20 km altitude) where we want to observe their vertical extent. Cloud turrets are therefore observed primarily with the camera pointing towards the limb so that the cloud extent can be resolved in altitude.
Scattered sunlight and light from lightning emissions will give the main cloud extent.

Thunderstorms are most frequent and powerful in the afternoon and early evening, typically from 16:00 – 22:00 LT. The pointing of the camera should therefore be primarily towards the ram-direction in order to avoid getting the sun in the FOV at the evening terminator. Once the ISS approaches the storm, the operator can deviate from limbpointing and select a cloud turret element to follow to 60 deg from nadir. The astronaut is expected to know enough about the data that are requested to make this judgement.

Ideally the observations will be conducted in two bands: in the near UV at 427.8 nm and in the red band at 630 nm. In practice, if these narrow bands are not available on the ISS, an alternative is to use available filters, in other words: one blue filter and one red filter, respectively. The bands will give us a measure of the distribution of ice crystals extending into the stratosphere, the cloud halo (Koren et al., 2007). On the night side, the illuminating source will be optical emissions from lightning reflected off clouds. The camera exposure time will depend on whether the main portion of the storm is on the dayor night side and on the camera system specification. A possible option could be a telescopic lens and a mechanical support. In any case sharpness of the images needs to be ensured.

Table 1: Cloud turrets characteristics

The expected photon flux at the instrument is shown in the table 1-3 assuming a 5 nm bandwidth of the filters. The characteristics are adapted from AD2, prepared by the ASIM Facility Science Team. Adaptation includes a change in the viewing angles (geometry) and inclusion of atmospheric absorption effects discussed inAD3. The transmission properties of the ISS windows are not included.

Gravity waves
Gravity waves will have wavelengths from 10- several 100 km and are observed in a thin airglow layer at 90-95 km altitude. The observation will give information on their horizontal extent, but are not expected to give information on the vertical extent. Gravity waves are therefore primarily observed at a slanted angle:0-30 deg from nadir. If 30 deg is not feasible 20 deg is also acceptable.

The emissions that we want to measure are at 630 nm but in practice, if this narrow band filter is not available on the ISS, an alternative is to use available filters, in other words: a red filter. A spatial resolution of at least 1 km or better is desired. Observations are only performed at the night side because scattered sunlight is too bright.

Table 2: Airglow characteristics

TLE observations are done in a broad spectral band with the camera pointing towards the limb, slanting 60-90 deg from nadir. A spatial resolution of 400 m or better is desired. The camera must run at 20 frames/s or more.

Table 3: TLE characteristics

The procedure follows these steps:
1. The Science Team on the ground will follow the global thunderstorm activity. A window in space and time for observations from the ISS of a target thunderstorm will be predicted 72 hours in advance and be communicated to the BUSOC. It is desired that the window can be refined as the target time approaches. Simultaneously, the Science Team will communicate the RO-number requested for the target and the specifications for the camera system.
2. The BUSOC notifies the science team if a requested target observation can be accommodated.
3. If the target observation can be accommodated the science team will provide 48 hours in advance the confirmation of the event occurrence. This confirmation will allow the start of the observation set-up.
4. If the target observations can be accommodated, the refined target information will be uplinked to the ISS not less than 24 hours in advance.
5. The astronaut mounts the camera system with the requested optics and implements the requested camera setting.
6. The astronaut remains with the camera system to take the requested images. It is expected that the astronaut will adjust the pointing to capture scientific events if they are off the estimated, uplinked target location.
7. The astronaut stops recording at the end of the time window.
8. The astronaut dismounts the camera system if needed.
9. Three images from the sequence should be down-liked to the science team within 3 hours or shorter (desired). This allows the science team to assess the instrument settings and adjust these if needed for the next observation sequence.
10. All data should be saved and made available for the project.
11. The science team should be notified within 3 hours (desired) if an observation sequence was not implemented.
Parameters Measured
Data measured from the ISS:
1. Optical camera pixel data (= image frames) of each observation sequence
2. One image above horizon (stars) for on-ground calibration of camera sensitivity
3. Time of observations with accuracy <10 ms (desired).
4. Pointing direction of the start of a sequence with an accuracy as good as possible (the goal is < 5 deg in the ISS reference system)

Other ISS data

1. Information on any deviations from the requested camera settings pointing or timing of the sequence.
2. ISS attitude at time of observation

Data from other sources
1. Global lightning activity data (e.g. WWLNN, GLD360 and other networks)
2. Cloud cover and cloud top temperatures from meteorological satellites

Mission Duration
Minimum Mission Duration: 8 days

Ground Reference Experiments
There are no ground reference experiments.

Inflight Session Requirements:
1. RO1-2:
a. Successful sequences: Required: 20, desired 50
b. Images/sequence: Required 3, desired 100

2. RO3:
a. Successful sequences: Required: 5, desired 20
b. Images/sequence: Required 500, desired 5000


The analysis proceeds in the following steps:
1. The image data will be calibrated using star fields.
2. The accuracy of the pointing direction will be improved by using optical ground features and lightning activity together with the location of the ISS.
3. From the image data, the science features and their intensities are extracted (science data).
4. The science data are correlated with information taken from other sources.
5. Scientific papers and reports will be written relating to the research objectives RO1-3.
Expected Results and Hypothesis

Initially the goal is that of a technical feasibility demonstration.
The experiment will demonstrate whether the choice of hardware and of operational requirements is adequate.
N.B. If these observations were to be repeated systematically over the long term (of the ISS lifetime) we would expect:
(1) to observe enough thunderstorms from the ISS to address the scientific questions with statistical significance. The data will allow for inputs into global atmospheric climate on water vapour transport in the atmosphere.
(2) to obtain enough data to be able to discriminate between models proposed for jets and gigantic jets. The required number of observations and data taken are given in subsection
"ASIM - Atmosphere-Space Interactions Monitor", http://www.asim.dk/.
A.N. Belyaev, V.V. Alpatov, E. Blanc, V.E. Melnikov, (2006), "Space-based observations of O2 A (0,0) band emission near the solar terminator and their interpretation", Advances in Space Research, 38, 11, pp. 2366-2373.
E. Blanc, T. Farges, R. Roche, D. Brebion, T. Hua, A. Labarthe, V. Melnikov, (2004), "Nadir observations of sprites from the International Space Station", Journal of Geophysical Research - Space Physics, 109, doi: 10.1029/2003JA009972, pp. A02306.
M.A. Geller, H.L. Liu, J.H. Richter, D. Wu, F.Q. Zhang, (2006), "Gravity Waves in Weather, Climate, and Atmospheric Chemistry: Issues and Challenges for the Community", Inspired by the June 2006 Gravity Wave Retreat hosted by The Institute for Integrative and Multidisciplinary Earth Studies (TIIMES) at the National Center for Atmospheric Research (NCAR) in Boulder, CO, U.S.A..
P.B. Hays, J.F. Kafkalidis, W.R. Skinner, R.G. Roble, (2003), "A global view of the molecular oxygen night glow", Journal of Geophysical Research - Atmosphere, 108, D20, doi: 10.1029/2003JD/003400, pp. 4646.
I. Koren, L.A. Remer, Y.J. Kaufman, Y. Rudich, J.V. Martins, (2007), "On the twilight zone between clouds and aerosols", Geophysical Research Letters, 34, L08805, doi:10.1029/2007GL029253.
B.E. Mapes, (1993), "Gregarious tropical convection", Journal of the Atmospheric Sciences, 50, 13, pp. 2026-2037.
S.B. Mende, G.R. Swenson, P. Geller, K.A. Spear, (1994), "Topside observation of gravity waves", Geophysical Research Letters, 21, DOI: 10.1029/94GL01696, pp. 2283-2286.
S.B. Mende, H. Frey, S.P. Geller, G.R. Swenson, (1998), "Gravity wave modulated airglow observation from spacecraft", Geophysical Research Letters, 25, DOI: 10.1029/97GL03224, pp. 757-760.
J.K. Nielsen, M. Foster, A. Heidinger, (2011), "Tropical stratospheric cloud climatology from the PATMOS-x dataset: An assessment of convective contributions to stratospheric water", Geophysical Research Letters, 38, doi: 10.1029/2011GL049429, pp. L18801.
T.A. Shaw, T.G. Shepherd, (2008), "Raising the roof", Nature Geoscience, 1, pp. 12-13.
S. Solomon, K.H. Rosenlof, R.W. Portmann, J.S. Daniel, S.M. Davis, T.J. Sanford, G.K. Plattner, (2010), "Contributions of stratospheric water vapor to decadal changes in the rate of global warming", Science, 327, 5970, doi:10.1126/science.1182488, pp. 1219-1223.
click on items to display

Figure 1: Thunderstorm clouds with turrets; observed from the ISS.

Figure 2: Cloud modulation by gravity waves generated by wind turbulence over the Sandwich Islands; observed from the ISS.

Figure 3: The air glow layer of the mesosphere modulated by gravity waves; observed from the ground with a fish-eye objective.

Figure 4: Transient Luminous Events (TLEs).

Figure 5: The annual lightning flash density; observed by the OTD and LIS instruments.

Table 1: Cloud turrets characteristics

Table 2: Airglow characteristics

Table 3: TLE characteristics
© 2019 European Space Agency