The Cyclone Intensity Measurements from the ISS (CyMISS) (TROPICAL CYCLONE) - 07.05.17

Overview | Description | Applications | Operations | Results | Publications | Imagery

ISS Science for Everyone

Science Objectives for Everyone
Tropical cyclones, called hurricanes or typhoons depending on the part of the world where they occur, cause more loss of life and property than any other natural phenomena on Earth. But most of the world is not covered by real-time information on these storms’ intensities, with the exception of North America and adjacent areas (such as the Bahamas and the Caribbean Islands). The Cyclone Intensity Measurements from the ISS (Tropical Cyclone) investigation demonstrates the feasibility of studying these powerful storms from space, which would be a major step toward alerting populations and governments around the world when a dangerous storm is approaching.
Science Results for Everyone
Information Pending

The following content was provided by Paul Joss, Ph.D., and is maintained in a database by the ISS Program Science Office.
Experiment Details

OpNom: Tropical Cyclone

Principal Investigator(s)
Paul Joss, Ph.D., MIT, Carlisle, MA, United States

Alva Stair, Visidyne, Burington, MA, United States
Gail E. Bingham, Ph.D., Utah State University, North Logan, UT, United States

Visidyne, Burlington, MA, United States

Sponsoring Space Agency
National Aeronautics and Space Administration (NASA)

Sponsoring Organization
National Laboratory (NL)

Research Benefits
Earth Benefits, Scientific Discovery

ISS Expedition Duration
September 2014 - March 2016; March 2016 - September 2017

Expeditions Assigned

Previous Missions
Information Pending

^ back to top

Experiment Description

Research Overview

  • Tropical cyclones, which are known by various names (hurricanes, typhoons, etc.) in different parts of the world, are the most devastating natural phenomenon on earth. In 1970, a single typhoon killed 600,000 people in Bangladesh. In 1900, a hurricane completely destroyed the city of Galveston, Texas and killed an estimated 10,000 people – the greatest loss of life in any natural disaster throughout the history of the United States. Hurricane Sandy (2012) caused $65 billion in property damage across the eastern United States, the greatest loss of property in a natural disaster throughout all of recorded history. 
  • Advance warning of the intensity of a tropical cyclone is crucial for the preservation of life and property. However, most of the world is unprotected by reliable measurements of the intensities of their storms before they make landfall. The exception is North America and adjacent landmasses (such as the Bahamas and the islands of the Caribbean), where U.S. hurricane hunter aircraft provide information on the intensities of hurricanes while they are still out to sea. However, these flights are expensive and dangerous (five aircraft and their crews have been lost in past years). Hurricane hunter flights are, therefore, carried out less frequently than would be needed to detect the very rapid changes in intensity that tropical cyclones often undergo. 
  • The physics of the key mechanisms driving tropical cyclones is well understood. By applying these known physical principles and using measurements routinely taken by Earth-orbiting meteorology satellites, the intensity of a strong tropical cyclone can be inferred to high accuracy. The only missing ingredient is measurement of the altitudes of the cloud tops within the eyewall (the region of extreme winds and torrential rainfall that lies just outside the cloud-free eye at the center of a tropical cyclone). 
  • The Tropical Cyclone Intensity Measurements from the ISS (CyMISS) project (also known as the Tropical Cyclone project) was selected by the Center for the Advancement of Science in Space (CASIS) to investigate filling in this missing link. It does so by measuring the cloud-top altitudes near the eye by applying pseudo-stereoscopy to sequences of photographic images, taken from the Cupola of the ISS, during ISS overpasses of these storms. The purpose of the project is to confirm the underlying physics of tropical cyclones, and to validate this new technique for measuring the intensities of tropical cyclones. Photographic measurements from the Cupola under controlled conditions (a fix-mounted camera with automated exposures) are the first step in establishing the feasibility of pseudo-stereoscopy for the altitude retrievals. The next step is to place an infrared camera/sounder with an integrated star sensor on the exterior of the ISS, to obtain coverage of tropical cyclones in darkness, as well as daylight, and to obtain the high angular accuracy required to meet the goal of measuring the eyewall cloud-top altitudes to within ~100 meters. When validated and implemented as a space-based system, this technique can be used to provide worldwide measurements of the intensities of strong tropical cyclones at frequent intervals.


The theoretical underpinnings of CyMISS (also known as Tropical Cyclone) are the widely accepted “Carnot engine” model for a tropical cyclone, developed mainly by K. Emanuel and his collaborators1. This model predicts that simultaneous measurement of the altitude and temperature of a storm's eyewall cloud tops, together with readily available sea surface temperatures and other ancillary data, can be used to determine the peak sustained winds within the storm. The investigation team expects this technique to be most accurate and reliable for well-developed tropical cyclones (TCs) with intensities of category 3 or higher (i.e., storms with sustained winds in excess of 180 km/hr).
In the underlying model, the circulation of air and moisture through the TC is accurately approximated by two uniform temperature (isothermal) flow regimes, and two adiabatic (without a gain or loss of heat) flow regimes. Each regime is equivalent to one stroke of a classical Carnot engine. In stroke 1, the air spirals from the storm’s outer circulation towards the center in a frictional boundary layer of ~1 km depth. Contact with the sea surface ensures that the flow is approximately isothermal, and large quantities of heat are added by evaporation from the sea. In stroke 2, the air ascends in the eyewall of the storm (the region of intense winds and torrential rainfall just outside the storm’s eye) and flows out to large distances (~2,000 km) from the storm center. This leg is nearly adiabatic, if the combination of dry air, water vapor, and condensed water is considered to be the working fluid. In stroke 3, the air descends in the lower stratosphere, while losing heat by radiation to space. This leg is nearly isothermal, owing to the isothermal structure of the ambient lower stratosphere. In stroke 4, the air descends back to the sea surface, while losing entropy by electromagnetic radiation to space and gaining entropy owing to mixing upward of water vapor by shallow cumulus clouds. Because the ambient atmosphere has a temperature lapse rate that is nearly moist adiabatic, the net entropy change on this leg is very nearly zero.
By applying the laws of thermodynamics to this model, one can derive a formula that determines the central sea-level pressure of the TC in terms of the ambient sea-level pressure on the periphery of the storm, the ambient sea surface temperature, and the altitude and temperature of the tops of the clouds above the eyewall. The ambient pressure and sea surface temperature are readily available from other sources, and the temperature of the cloud tops can be measured to the requisite level of accuracy (±1 K) with the new class of GOES-R and Himawari 8/9 meteorology satellites. A propagation-of-errors analysis demonstrates that if the remaining parameter, the altitude of the eyewall cloud tops (ECTs), can be measured to an accuracy of ±100 meters, then the central pressure of the TC can be determined to ±3.5 millibars. This is far more accurate than can be achieved with other remote-sensing techniques; only in situ measurements with dropsondes from hurricane hunter aircraft have a higher level of accuracy. Measurement of the central pressure to an accuracy of ±3.5 millibars is sufficient to determine the peak sustained winds in a well developed TC to an accuracy of ±10 mph (90% confidence). Knowledge of the real-time central pressure of a TC is also a key ingredient in the initialization of the numerical weather prediction codes that are used to forecast the future surface track and intensity changes of the storm.
In this method of measurement of TC intensities, the most difficult parameter to measure is the altitude of the ECTs. CyMISS (Tropical Cyclone Measurements from the ISS) is a proof-of-concept project intended to validate this method and demonstrate a viable technique for accurately measuring the altitudes of the ECTs from low-Earth orbit. As the ISS passes a TC at distances up to ~2,000 km, the ISS crew takes a series of between 100 and 240 photographic images of the ECTs at 1-second intervals. The ~8 km/s forward motion of the ISS relative to the earth’s surface, during the ~1-4 minutes that a given TC remains within range for sufficiently high-resolution imagery, provides a baseline of up to ~2,000 km for each series of images. Image rectification and stereoscopic techniques are employed to these images in order to deduce the cloud-top altitudes to the requisite accuracy. Analysis of the imagery must also take into account the rapid motions (up to ~300 km/hr) of the clouds themselves. As a result, the analysis also yields measurements of the high-altitude winds in which the clouds are embedded, providing another useful measure of the intensity of the TC.
If this project is successful, a future phase of CyMISS is to be deployed on the exterior of the ISS, with an infrared camera/sounder that can image TCs in darkness as well daylight, and can also measure the temperatures of the ECTs without the need to rely on simultaneous data collection by independent meteorological satellites. Co-aligned star cameras are also deployed in this phase in order to improve the precision of the stereoscopic measurements of ECT altitudes. A future operational system, comprising a constellation four nanosatellites (such as Cubesats), each with the same instrument suite, could provide continuous worldwide coverage of the world’s oceans. Such a system could furnish nearly real-time measurements of the intensities of all intense tropical cyclones, with an average interval of only ~1.5 hours between successive measurements of any given storm.
1Emanuel, K. A. 2003, Annual Review of Earth and Planetary Sciences, Vol. 31, pp. 75–104, and references therein.

^ back to top


Space Applications
Advance warning of a tropical cyclone’s intensity is crucial for protecting lives and property in the face of a storm. But hurricane-hunting aircraft only cover a small portion of the planet affected by such storms. Monitoring storm strength from space is safer, less expensive, and potentially more effective with a new meteorological technique that uses photographs taken from the International Space Station’s Cupola to measure the altitudes of cloud tops near the storm’s eye. The altitude of these clouds is related to the intensity of a strong tropical cyclone. Results from this investigation could validate a new space-based system for measuring the intensity of dangerous tropical storms.

Earth Applications
In 1970, a single typhoon killed 600,000 people in Bangladesh. In 2012, Hurricane Sandy caused $65 billion in property damage across the eastern United States, the greatest loss of property in a natural disaster in recorded history. But better advance warning allows people time to prepare and to evacuate. A space-based system that monitors storm intensities around the world can greatly reduce loss of life and property in these powerful and extremely dangerous storms.

^ back to top


Operational Requirements and Protocols

The objective of the CyMISS (also known as Tropical Cyclone) project is to demonstrate the feasibility of measuring the altitudes of the cloud tops in tropical cyclones (TCs) from low-Earth orbit. This is accomplished by applying pseudo-stereoscopy to sequences of high-resolution images taken by ISS crew members during overflights of these storms. To accomplish this objective, the crew is requested to following the following protocols:
Visidyne, Inc. notifies NASA-JSC in a timely manner of any upcoming ISS overflights of a targeted TC, including the predicted times of the overflights, and whether the ground track of the ISS will pass to the left or right of the TC on each overflight. These data are transmitted to the ISS via Crew Earth Observation (CEO) message or Execute Note.
All imagery is to be taken with the onboard Nikon D4S camera fitted with the onboard Nikkor 50mm fixed length lens, with a battery and flash memory card installed within the camera. The camera is to be mounted within the Cupola with a multi-use bracket, and oriented to view out of either Window 1 or Window 4, depending on whether the ground track of the ISS is predicted to pass to the left or right of the TC. The camera should be (1) oriented so that the short axis of its field of view is parallel to the Earth’s horizon (so that the TC passes through the short axis of the image), and (2) aimed slightly upward, so that both the horizon and the center (eye) of the TC are within the field of view at the time of closest approach of the TC to the ISS. When properly set, the field of view should be similar to that shown in Figure 1, except that the short axis of the field of view should be parallel to the horizon. The Nikkor lens should be set and locked at its minimum aperture, and the lens focus should be set to infinity (∞).
The camera body should be displayed, or be fixed, at the following settings:
  • The power switch should be ‘ON’.
  • The top LCD should indicate that both the battery level and the number of frames remaining are sufficient.
  • Exposure Compensation: ‘0.0’
  • Camera Mode: ‘M’
  • Shutter Speed: ‘2000’
  • f/stop: ‘11’
  • Diopter: Adjust as preferred by crew member
  • Light Meter: matrix mode
  • Bracketing (BKT button): '0F'
  • Frame Rate: ‘S’
  • The Body Focus Mode should be switched to ‘M’.
  • Rear LCD settings should be set to \ ISO = '400', QUAL = 'RAW', and WB = 'A'.
  • The camera’s flash should be powered off or not installed.
The date and time settings of the camera are to be established by pressing the camera’s MENU button and then using the Navigate Pad to sequentially select ‘SETUP MENU’, ‘Time zone and date’, and ‘Date and time’. The Navigate Pad is then to be used to set the date and time to the correct Greenwich Mean Time (GMT), in the format 20YY/MM/DD HH:MIN:SEC. The date and time entry are completed by choosing ‘select’ on the Navigate Pad, pressing the OK button, and then pressing the MENU button twice. (A desired field is chosen by pressing up or down on the Navigate Pad; the field can then be selected by pressing either left or right on the pad.)
Obtaining sequences of images at intervals of precisely one second is critically important for the success of the project. Such sequences can be obtained by use of the camera’s built-in intervalometer. The desired intervalometer settings are chosen by use of a procedure similar to that for setting the correct time:
For crew-tended imagery, the intervalometer is set by first pressing then MENU button and then using the Navigate Pad to choose the desired settings. Next, ‘SHOOTING MENU’ followed by ‘Interval timer shooting’ are selected. (If ‘In progress’ [‘Pause’] is displayed, ‘Off’ should be selected and the OK button pressed, and ‘SHOOTING MENU’ followed by ‘Interval timer shooting’ should again be selected.) Next, the ‘Choose start time’ option, ‘NOW’, and ‘Interval’ are to be selected sequentially, and ‘Interval’ is to be set to ‘00:00:01’. ‘No. of times x no. of shots’ is now be selected, with the number of intervals and the number of shots then set to ‘100’ and ‘1’, respectively. Finally, when the central portions of the TC approach the edge of the camera’s field of view, the ‘Start’ and ‘On’ options are to be sequentially selected by use of the Navigate Pad, and the OK button is to be pressed in order to begin photographing the sequence of images.
For imagery that is not tended by a crew member, a somewhat different intervalometer setup procedure is to be followed. The first portion is identical to that for tended imagery: First, ‘SHOOTING MENU’ and then ‘Interval timer shooting’ should be selected. (If ‘In progress’ [‘Pause’] is displayed, ‘Off’ should be selected and the OK button pressed, and ‘SHOOTING MENU’ followed by ‘Interval timer shooting’ should again be selected.) Next, the ‘Choose start time’ option is to be selected. The remainder of the procedure differs from that for crew-tended imagery. The ‘Choose start time’ and ‘Start time’ options are to be selected sequentially, and the start time is to be set at two minutes prior to the value given in the CEO Message or Execute Note. ‘Interval’ is then to be set to ‘00:00:01’. ‘No. of times x no. of shots’ is next be selected, with the number of intervals and number of shots set to ‘240’ and ‘1’, respectively. (The larger number of shots, compared to 100 shots for crew-tended imagery, compensates for any errors in the predicted location of the TC, and thereby ensures that the targeted TC is captured within the image sequence.) Finally, the ‘Start’ and ‘On’ options are selected sequentially by use of the Navigate Pad, and the OK button must be pressed. Photography of the image sequence commences automatically at the selected time.
For both crew-tended an untended imagery, it should be verified that the lens focus remains set to infinity (∞) and the end of the camera setup procedure.
After the image sequence is completed, the camera is to be dismounted. The camera should then be deactivated by setting the camera’s Mode to ‘M’ and switching the Body Focus mode to ‘S’.
The sequence of images is downloaded when convenient, and should be labeled with the name of the TC, as specified in the CEO Message or Execute Note.

^ back to top

Decadal Survey Recommendations

Information Pending

^ back to top

Results/More Information

Information Pending

^ back to top

Results Publications

^ back to top

Ground Based Results Publications

^ back to top

ISS Patents

^ back to top

Related Publications

    Latvakoski H, Bingham GE, Topham TS, Podolski IG.  Phase change cells and the verification of gallium as a thermal calibration reference in space. Infrared Remote Sensing and Instrumentation XXIII, San Diego, California; 2015 September 22 96080Q.

    Topham TS, Bingham GE, Latvakoski H, Podolski IG, Sychev VN, Burdakin A.  Observational study: microgravity testing of a phase-change reference on the International Space Station. npj Microgravity. 2015 August 20; 1: 15009. DOI: 10.1038/npjmgrav.2015.9.

^ back to top

Related Websites

^ back to top


Hurricane Edouard (ISS041-E-14067) over the Atlantic Ocean, taken from the ISS on September 16, 2014 at 13:51:49 GMT (8:51:39 am Central Daylight Time). At that time, Edouard was a category 2 (moderately strong) hurricane.  Image courtesy of Earth Science and Remote Sensing Unit, NASA-JSC.

+ View Larger Image

A montage of twelve images, each covering an area of 62 x 62 miles centered on the cloud-free eye of Edouard, taken from the ISS over the course of an 82-second interval. The image in the fourth column and second row of the montage was derived from the large-scale image of Edouard shown in Figure 1. The time stamp on each image gives the time (in GMT) that each image was taken, expressed as hours:minutes:seconds. The bright “eyewall” clouds lie just outside the eye and whirl around it at speeds of up to 110 mph (up to 230 mph in the most intense hurricanes). The tops of these clouds lay 30 – 50,000 ft. above sea level.  Image courtesy of Visidyne/NASA-JSC.

+ View Larger Image