Measuring the Ocean and Air-Sea Interactions with Unmanned Aerial Vehicles

Benjamin D. Reineman

At the time of the first ALPS meeting in 2003, unmanned aerial vehicles (UAVs),1 though already a staple for military surveillance, were out of reach for much of the oceanographic community. Lower costs and improved positioning, control, and ease of use have since opened doors for scientists with less flight expertise and a more modest budget. Technology has improved such that aircraft can launch and recover from a modestly sized research vessel, either by net or catch lines, or, in the case of a small multi-rotor craft, even in the palm of the hand.

Unmanned aircraft for ocean-related science is a growing and multi-faceted field, with platforms and field campaigns ranging in scales from week-long missions with NASA-operated 130-ft wingspan GlobalHawks outfitted with weather radar and dozens of dropsondes, down to missions of tens of minutes with commercial off-the-shelf multi-rotor craft and a camera. This report is an attempt to brief the oceanographic community on the current state of the art in oceanographic science enabled by unmanned aircraft and comment on their potential future in the field. Figure 1 presents a sampling of various UAVs used in oceanographic research.

Figure 1. Examples of unmanned aerial vehicles presently used for oceanographic or atmospheric research, with sample studies referenced.


With the exception of a number of high-altitude, solar-​powered prototype crafts (notably “pseudo-satellite” efforts by NASA, Facebook, Google, and others as high-altitude communication nodes), petroleum-based fuels are still the preferred energy source for endurance-focused UAVs. As with many instruments in oceanography, a major advance in battery technology will open up many new opportunities. At present, we are often still bound to gasoline, which has 50 to 100 times the specific energy density of commercially available lithium ion cells.2 For mid-size (20 kg) fixed-wing craft, electric planes can typically stay aloft for a few hours, while gasoline-powered planes can perform missions up to 24 hours (with a trade-off of payload and additional fuel).

Small multi-rotor craft have seen incredible commercial popularity growth in the last five years, driven by amateur and professional videographer demand. Given their small size, relatively low price, and pinpoint maneuverability and stability, they are an attractive alternative to fixed-wing UAVs, when limited endurance and range are not restricting factors. They are usually a few kilograms or less, with payload capacities of a few hundred grams, and endurances of 20 to 30 minutes (powered by rechargeable lithium batteries). For fine-scale atmospheric measurements, the propeller wash is a potential issue, though some studies have investigated multi-rotor craft for atmospheric sampling (e.g., Machado, 2015).

Land-based Earth and atmospheric research with UAVs is more well established than that over the ocean, given more straightforward access to runways for launch and recovery (typically required by medium and large fixed-wing craft). Aviation restrictions have historically hindered ocean and marine atmospheric boundary layer studies from land-launched UAVs, as these missions required approved corridors to sanctioned ocean airspace, but recently updated aviation regulations have opened up more airspace. Additionally, in recent years, fixed-wing craft have pursued innovative launch and capture techniques, or VTOL (vertical take-off and landing), which have and will enable expanded oceanographic, air-sea interaction, and marine atmospheric boundary layer research.

To date, a large portion of the science conducted with unmanned vehicles has been imagery-based, using small commercially available platforms. For under $1000, a quadcopter capable of carrying a high-definition camera that can stream imagery back to the ground control station, which in some cases is just a smartphone or tablet, can be acquired. Marine surveillance and situational awareness have been strong drivers of maritime UAV use. A recent chapter in the Handbook of Unmanned Aerial Vehicles by de Sousa et al. (2014) reviews thoroughly the state of UAVs for maritime operations, including search and rescue, ice operations, and coastal and shipping security. In the scientific community, early adopters of UAV imaging have been marine mammal surveyors (Durban et al., 2015), where cost-effective cetacean and pinniped surveys can be performed with minimal behavioral disturbance.

Infrared imaging from UAVs has permitted small- to mesoscale observations of surface temperature structure. Using thermal imaging aboard ScanEagles, Zappa et al. (2013) and Maslanik (2016) examined surface meltwater from sea ice, and Reineman et al. (2013) examined Langmuir-type circulations aligned with the wind (Figure 2a). Upcoming experiments using smaller multi-rotor craft with thermal imaging hope to examine fine- and mesoscale surface temperature structure (Figure 2b), crucial to understanding and modeling air-sea interaction.

Figure 2. (a) Sample infrared imagery from a FLIR A325 aboard a ScanEagle (adapted from Reineman et al., 2013) showing Langmuir-like surface signatures aligned with the wind. (b) Uncalibrated test image from a FLIR-DJI ZenmuseXT aboard a small DJI Inspire 1 quadcopter (inset) showing a temperature front (A.F. Waterhouse, E. Lo, and D. Rissolo, pers. comm.).

High-resolution wavefield measurements are important for air-sea interaction research and wave modeling, and are intriguing for satellite altimetry calibration and validation of “sea-state bias” (Melville et al., 2016). From a ScanEagle, single-point lidar for along-track surface elevation measurements was demonstrated for surface wave measurements (Reineman et al., 2013) as well as for surface signatures of internal waves (Reineman et al., 2016). While state-of-the-art complete scanning lidar acquisition and imaging packages are in the 20–50 kg range, smaller packages may facilitate this technology to transition to the unmanned aircraft realm. RIEGL (Austria; now has a commercially available, fully outfitted multi-rotor craft with scanning lidar (RiCOPTER), a 25 kg electric craft with endurance up to 30 minutes. Technology such as this will greatly expand the sampling capability of ocean surface waves, and if deployed from a research vessel, will provide accurate surface wave measurements over any ocean region.

UAVs used for standard atmospheric soundings have been employed for over a decade in the marine atmospheric boundary layer. Mean winds can be inferred by comparing airspeed and heading to GPS-derived ground speed and course over ground. Combined wind, temperature, and humidity measurements over a spatial distribution can be used to quantify bulk heat fluxes. Since 2009, Knuth and Cassano (2014) and Cassano et al. (2016) have been flying routinely in the Antarctic, measuring, among other things, the polynyas coming down the West Antarctic Ice Sheet. Bradley et al. (2015) and Zappa (2016) are experimenting with UAV-launched, air-deployed microbuoys for atmospheric soundings and also Lagrangian surface-layer temperature measurements. UAV atmospheric data have also been assimilated into real-time coupled ocean-atmosphere models in a manner similar to balloon-sonde data (Doyle et al., 2016; Reineman et al., 2016). The advantages of UAV atmospheric profiles over balloon profiles include reusability, directed and reproducible tracks, and sampling of horizontal gradients.

Figure 3. Vertical profiles of momentum flux in the marine atmospheric boundary layer, as measured by a turbulence probe on a ScanEagle. From Reineman et al. (2016)

For three-dimensional, high-resolution turbulent wind measurements, which are necessary for directly measuring turbulent air-sea heat and momentum fluxes (using eddy-covariance techniques), multi-port pressure probes have been developed and combined with high-accuracy inertial and GPS units. Such a sensor has been implemented on a ship-launched Boeing-Insitu ScanEagle, measuring momentum flux, and latent and sensible heat fluxes in the marine atmospheric boundary layer during several field campaigns (Reineman et al., 2016). Figure 3 presents vertical profiles of momentum flux as measured during cross- and along-wind segments, where the differences in fluxes between cross- and along-wind are attributed to the presence of planetary boundary rolls. The low altitude required for accurate air-sea fluxes (typically 30 m) is below the typical limit for safe manned aircraft operation.

The Federal Aviation Administration (FAA) is also embracing UAV technology. With new regulations issued in July 2016, flights in general airspace are permitted for UAVs below 55 pounds, following some basic rules, including but not limited to: staying below 400 ft, staying away from populated areas, and maintaining visual line-of-sight. Easements of these rules and others can be obtained through a straightforward process. The pilot-in-command must have passed an online certification course.3 These new regulations will surely permit increased access to oceanographic studies with UAVs in coming years.

Unmanned aircraft are primed to bring the next wave of oceanographic, marine atmospheric boundary layer, and air-sea interaction measurements to scientists’ desks. They have the ability to fly dangerous missions at little risk to human operators, or to fly long-endurance or tedious missions, giving novel measurements of the atmosphere or ocean surface. The immense range of scales in sensor and platform cost and complexity results in a wide range of scales of physical processes that can be measured and questions that can be answered. When combined with shipboard sampling, unique space and time data sets from the subsurface up into the atmosphere can be generated, and point measurements from the ship can be placed in a larger atmospheric and oceanographic context.



1. I give preference here to the term UAV rather than UAS (unmanned aerial system; though neither really ought to be gender-specific), which refers to the platform along with the ground station and any associated infrastructure. The term “drone” is avoided as it has a military connotation, and can refer to missiles as well

2. If we consider drivetrain efficiency of electric systems to be much more efficient than internal combustion (lighter comparable engines and much more efficient energy conversion), the available power output per kg storage medium for a complete gasoline system is closer to 5 to 20 times that for an electric system, but there are still many trade-offs to consider.

3. For the full text, see (or search “FAA Part 107”).



Albertson, R., S. Schoenung, M. Fladeland, F. Cutler, and B. Tagg. 2015. Enabling Earth science measurements with NASA UAS capabilities. The International Archives of Photogrammetry, Remote Sensing and Spatial Information Sciences 40(7):1,111–1,117,

Bradley, A.C., S. Palo, G. LoDolce, D. Weibel, and D. Lawrence. 2015. Air-deployed microbuoy measurement of temperatures in the marginal ice zone upper ocean during the MIZOPEX campaign. Journal of Atmospheric and Oceanic Technology 32(5):1,058–1,070,

Cassano, J.J., M.W. Seefeldt, S. Palo, S.L. Knuth, A.C. Bradley, P.D. Herrman, P.A. Kernebone, and N.J. Logan. 2016. Observations of the atmosphere and surface state over Terra Nova Bay, Antarctica, using unmanned aerial systems. Earth System Science Data 8:115–126,

de Sousa, J., P. McGuillivary, J. Vicente, M. Nunes Bento, J.A.P. Morgado, M. Madruga Matos, R. Ayres Gomes Bencatel, and P. Mónica de Oliveira. 2014. Unmanned aircraft systems for maritime operations. Pp. 2,787–2,811 in Handbook of Unmanned Aerial Vehicles. K.P. Valavanis and G.J. Vachtsevanos, eds, Springer,​10.1007/978-90-481-9707-1_75.

Doyle, J.D., T.R. Hold, and D.D. Flagg, eds. 2016. Trident Warrior 2013. NRL Technical Memo, PAO 16-1231-2460, 241 pp.

Durban, J.W., H. Fearnbach, L.G. Barrett-Lennard, W.L. Perryman, and D.J. Leroi. 2015. Photogrammetry of killer whales using a small hexacopter launched at sea. Journal of Unmanned Vehicle Systems 3(3):131–135.

Knuth, S., and J. Cassano. 2014. Estimating sensible and latent heat fluxes using the integral method from in situ aircraft measurements. Journal of Atmospheric and Oceanic Technology 31:1,964–1,981,

Machado, C.R. 2015. An Analysis of Meteorological Measurements Using a Miniature Quad-Rotor Unmanned Aerial System. Master’s thesis. Naval Postgraduate School, Monterey, California, 79 pp.

Maslanik, J.A. 2016. Investigations of Spatial and Temporal Variability of Ocean and Ice Conditions In and Near the Marginal Ice Zone: The “Marginal Ice Zone Observations and Processes Experiment” (MIZOPEX) Final Campaign Summary. Technical Report, DOE DOE/SC-ARM–15-046, 55 pp.

Melville, W.K., L. Lenain, D.R. Cayan, M. Kahru, J.P. Kleissl, P.F. Linden, and N.M. Statom. 2016. The modular aerial sensing system. Journal of Atmospheric and Oceanic Technology 33(6):1,169–1,184,

Reineman, B.D., L. Lenain, N.M. Statom, and W.K. Melville. 2013. Development and testing of instrumentation for UAV-based flux measurements within terrestrial and marine atmospheric boundary layers. Journal of Atmospheric and Oceanic Technology 30:1,295–1,319,

Reineman, B.D., L. Lenain, and W.K. Melville. 2016. The use of ship-launched fixed-wing UAVs for measuring the marine atmospheric boundary layer and ocean surface processes. Journal of Atmospheric and Oceanic Technology 33(9):2,029–2,052,

Zappa, C., S. Brown, W.J. Emery, J. Adler, G.A. Wick, M. Steele, S.W. Palo, G. Walker, and J.A. Maslanik. 2013. Local effects of ice floes on skin sea surface temperature in the marginal ice zone. Paper presented at the 2013 fall meeting of the American Geophysical Union, San Francisco, California, Abstract OS14A-04.

Zappa, C. 2016. UAS activities at Lamont-Doherty Earth Observatory of Columbia University. Presentation at the 2016 SCOAR Meeting, Marina, California, University-National Oceanographic Laboratory System, Scientific Committee for Oceanographic Aircraft Research.



Benjamin D. Reineman, Scripps Institution of Oceanography, University of California, San Diego, La Jolla, CA, USA,