Our Research


Space is not far away. Only 30 km above Earth’s surface—less than the length of a marathon race—the blue sky fades to black and stars pop out in broad daylight. At such altitudes, the ambient temperature (as low as -60 C), air pressure (<1% sea level), and cosmic radiation environment are akin to the planet Mars.
In 2011, a group of student researchers in Bishop, California, began launching sub-orbital helium balloons to explore this near-space environment. Their organization, “Earth to Sky Calculus,” has since flown more than 85 missions to the stratosphere for a fraction of the cost of traditional space missions. Many of the flights have tested technologies for a “Space Weather Buoy” to measure the response of the upper atmosphere to solar storms.
The Buoy is now fully developed and here we report findings from flights in the past year. Key results include the detection of two Forbush Decreases in 2014 and a surge in upper atmospheric radiation during the St. Patrick’s Day Storm of March 2015. The technology of the Buoy is affordable and accessible to high schools and small colleges, and could form the basis for a continental network of space weather balloon programs.


The Space Weather Buoy is a lunch-box sized payload that can measure temperature, ionizing radiation, and GPS altitude from sea level to more than 120,000 feet.   The payload ascends to the stratosphere, carried by a helium-filled balloon, and descends by parachute after the balloon pops.
The payload weighs approximately 2 kilograms. It is not only “lunch-box sized,” it actually is an insulated lunchbox (CoolCellTM), which provides +30 C of thermal protection to the interior. Instrumentation is mounted on both the interior and exterior surfaces of the container.
To lift the payload, we use a 1600 gm balloon made of natural rubber, inflated to approximately 8 feet diameter by 200 cubic feet of helium. A 10-foot diameter parachute designed for model rocketry is tied between the payload and the balloon. This payload stack ascends to the stratosphere at a near-constant velocity of 5.5 m/s (1100 ft/min). The velocity of the descent is not constant because of varying drag as the parachute sinks into the thickening atmosphere. Typical landing speeds are near 8 m/s (1600 ft/min).

The following instruments are mounted on the payload:

  1. Polimaster PM1621M radiation sensor: This device, mounted inside, senses X-rays and gamma-rays in the energy range 10keV to 20 MeV. Early versions of the Space Weather Buoy carried only one PM1621M. We now routinely fly four of the devices for redundancy and to improve the signal-to-noise ratio of our measurements. Each sensor contains a datalogger, which records dose rates in units of microRads/hr every minute during the flight. (cost per sensor $900)
  2. TinyTag TV-4204 cryogenic thermometer: This device consists of a metal probe mounted outside the payload, and a temperature logger, mounted inside, that records the exterior temperature every 1 second during the flight. ($250)
  3. SPOT Trackers: Two standard GPS hiking trackers are mounted on orthogonal faces of the payload’s exterior. They provide coordinates as often as every 2 minutes during the balloon’s flight, which we monitor in realtime using SPOT LLC’s online interface (http://findmespot.com). When the balloon parachutes back to Earth, signals from the SPOT trackers guide us to the landing site. The orthogonal mounting of the trackers insures that no more than one tracker can land face-down in the dirt. ($150 per tracker)
  4. GPS altimeter #1: Mounted to the outside of the payload, a Garman eTrex20 GPS tracker records the balloon’s 3D coordinates every 10s during the flight up to altitudes as high as 98,422 feet. ($200)
  5. GPS altimeter #2: A recent addition to the payload, this experimental altimeter, based on the Arduino MEGA microprocessor, records the balloon’s 3D coordinates every 1 second during the flight up to altitudes as high as 185,000 ft. The circuitry of device is mounted inside the payload bay; the antenna is mounted outside. ($200)
  6. Video cameras: Two GoPro Hero3+ video cameras are mounted inside the payload, looking out of camera bays to film the sky and landscape during the flight. A typical flight produces 64 GB of HD video. ($500 per camera)
The total cost of this payload is $5550. It is relatively easy to assemble, and is regularly launched by high-school teams unassisted by adults.
The “Edge-of-Space Port” where we release our balloons is located at an altitude of 8500 ft in the Eastern Sierra mountain range near Bishop, California. Landing sites are typically located 10 miles to 70 miles away in the Death Valley National Park, White Mountains, and the Inyo Mountain range of California.


Figure 1 shows a sample profile of ionizing radiation collected during a 2.5 hour flight to the stratosphere on May 30, 2015. The maximum altitude of the flight was 111,200 feet.


Figure 1: Ionizing radiation dose rates measured during a May 30, 2015, flight of an Earth to Sky Calculus “Space Weather Buoy.” The curve is an average of data collected by four identical PM1621 radiation sensors, which respond to X-rays and gamma-rays in the energy range 10keV to 20 MeV.
Newcomers to the field of atmospheric radiation measurements often expect dose rates to peak at the apex of the flight. Instead, the peak occurs approximately halfway up, in the lower stratosphere. When cosmic rays crash into Earth’s atmosphere, they produce a spray of secondary particles. With increasing depth in the atmosphere, the primary cosmic radiation component decreases, whereas the secondary radiation component increases. This situation results in a maximum dose rate at an altitude of ~20 km, the so-called “Pfotzer maximum,” named after physicist George Pfotzer who discovered the peak using balloons and Geiger tubes in the 1930s.
We have radiation profiles similar to this one on 37 dates between Oct. 27, 2013 and June 8, 2015. Regular flights to the stratosphere have allowed us to draw some conclusions about the average width and altitude of the Pfotzer Maximum, and to sample the stratosphere during multiple solar storms.
Because the Pfotzer Maximum is broad, our balloons spend a significant amount of time inside it, typically 50 minutes or more per flight. This gives us long integration times and high signal-to-noise ratios. The peak of each Pfotzer Maximum is found by fitting a simple parabolic curve to 1-minute averaged data. Typical values are near 400 uR/hr with uncertainties of order 1%. The average width of the Pfotzer Maximum, defined as the full width at 90% of the peak, is 25,600 feet +/- 2,600 feet. The average altitude of the peak is 66,900 feet +/- 1,400 ft. The quoted uncertainties are 2 standard deviations wide.
The goal of our monitoring program is to detect changes in the upper atmosphere in response to solar activity. We have succeeded on three occasions, summarized below.
The first case was Sept. 12, 2014.  A pair of CMEs hit Earth’s magnetic field, sparking a G3-class (Kp=7) geomagnetic storm. We launched a balloon during the peak of the storm, naively expecting to observe an increase in ionizing radiation compared to earlier flights during quiet conditions. Instead, we detected a sharp decrease, shown in Figure 2.


FIGURE 2: Each point in this plot shows the peak of the Pfotzer Maximum measured during a flight of Earth to Sky’s Space Weather Buoy. Error bars are estimated from the scatter of measurements by independent sensors onboard each payload.
This effect is called a “Forbush Decrease,” named after physicist Scott E. Forbush who first described it in the 20th century. Magnetic fields in the CMEs scattered cosmic rays away from Earth, causing a decrease in the peak of the Pfotzer Maximum. The peak on Sept. 12th dropped approximately 9% compared to flights before and after Sept. 12th.
This Forbush decrease was also observed by neutron monitors around the world such as the Oulu Cosmic Ray Station in Finland and the global network of stations operated by the Bartol Research Institute. Ground-level neutron counts are considered to be a good proxy for cosmic rays. The timing and magnitude of the Forbush Decrease we measured in the stratosphere over California were in good accord with ground-level neutron data. As an example, Figure 3 compares our data to those of Oulu.



FIGURE 3: Ionizing radiation measurements in the stratosphere over California (below) compared to ground level neutron counts from Oulu, Finland (above). Qualitatively, the two data sets are in good accord despite wide differences in altitude and latitude.
We observed another Forbush Decrease in late December 2014. On Dec. 21-22, a pair of CMEs delivered glancing blows to Earth’s magnetic field. The result was a minor G1-class (Kp=5) geomagnetic storm. The geomagnetic storm subsided in a matter of hours, but the associated Forbush Decrease lasted for nearly one month. Figure 4 shows neutron counts from Oulu, Finland.
During this slowly evolving event, we launched three balloons, on Dec. 24, 2015; Jan. 10, 2015; and Jan. 14, 2015. Blue dots in Figure 4 denote the peak of the Pfotzer Maximum, which we measured in the stratosphere, expressed as a percentage of the 2014 annual average.
For this particular event, balloon data were tightly correlated with the Oulu neutron counts—even more so than during the Sept. 12th Forbush Decrease. The degree of correlation is surprising given the geographic separation between observing sites and the fact that the two data sets represents such different ways of sampling the cosmic ray environment.


FIGURE 4: The slow Forbush Decrease of Dec. 2014-Jan. 2015. The red curve traces neutron counts from the Cosmic Ray Station at Oulu, Finland. The blue dots are measurements of the Pfotzer Maximum in the stratosphere over California. To make a comparison possible, both data sets are expressed as a percentage change with respect to a long-term average.

The third case occurred on Mar. 17, 2015, when a CME struck Earth’s magnetic field, triggering a severe G4-class (Kp=8) geomagnetic storm, at the time the strongest geomagnetic storm of Solar Cycle 24. We explored the “St. Patrick’s Day Storm” with a series of five balloon launches—one before the storm (March 13), one during the storm (March 17), and three after the storm (March 21, March 24, and April 1, 2015).
Figure 5a shows the data from those five flights. Instead of a Forbush Decrease, we measured an increase in ionizing radiation. This could be a result of the geomagnetic cutoff rigidity at our location changing during the severe geomagnetic storm. Energetic particles normally trapped around Earth’s polar regions migrated to lower latitudes—a ‘spillage’ which did not occur during the lesser storms of Sept. 12, 2014 and Dec. 22, 2014.


Figure 5b compares our stratospheric balloon measurements to Oulu neutron counts. The two trends are opposite. The ground-level neutron monitor observed a decrease in radiation, while the balloons observed an increase. This is qualitatively consistent with the idea of energetic particles migrating to lower latitudes during the severe storm.



We have succeeded in measuring changes to the radiation environment of the stratosphere during three solar and geomagnetic storms. These data are of interest to atmospheric scientists, space weather forecasters, and entrepreneurs of space tourism who hope to build vessels to carry human passengers to the same altitudes as our balloons.
There is, however, much room for improvement in our radiation payload. For one thing, it only samples a fraction of the total radiation environment. For example, our payload is completely insensitive to neutrons. Neutrons are a very important form of cosmic radiation, providing at least half of the biologically effective radiation dose at altitudes of interest to aviation and space tourism. To monitor this type of radiation, we are currently experimenting with bubble chambers and active neutron dosimeters that can fly successfully onboard a small balloon payload.
We would also welcome collaboration with more senior groups who occasionally fly sophisticated radiation sensors onboard research aircraft and heavy-lift balloons. Including our payload as a “hitchhiker” on such flights would allow a valuable cross calibration of our low-energy sensors with devices of greater sensitivity and spectral range.
Finally, we wish to form a “Space Weather Balloon” network, geographically distributed across North America, and ultimately across the globe. Simultaneous launches of payloads like the one we have developed could reveal the 3D response of the atmosphere to solar storms over wide areas, not just in a single flight path over California.


We have shown that it is possible to measure dynamics in upper atmospheric radiation using a relatively inexpensive payload for small helium balloons. The price point is within easy reach of many high schools and small colleges, and the technological know-how is not prohibitive. Most of the devices contained inside the lunch-box payload are off-the-shelf and do not require advanced degrees in electrical engineering or radiation physics to operate.
Despite the relative simplicity of the payload—or perhaps because of it—Space Weather Buoys are able to make meaningful astrophysical measurements. In less than one year, we have detected changes in the stratosphere in response to three solar storms. In two cases (Sept. and Dec. 2014), Forbush Decreases in the stratosphere were well-correlated with similar Forbush Decreases observed by neutron monitors on the ground. During the St. Patrick’s Day Storm of 2015, however, ground measurements of neutrons and balloon measurements of ionizing radiation were almost perfectly anti-correlated. This shows that we have a lot to learn, and using our techniques, young scientists are fully capable of participating in this exciting research.