How Unmanned Aircraft Systems (UAS, aka “drones”) are being applied in conservation research

By Leila Lemos, Ph.D. Student, Department of Fisheries and Wildlife, OSU

 

Unmanned Aircraft Systems (UAS), also known as “drones”, have been increasingly used in many diverse areas. Concerning field research, the use of drones has brought about reduced errors, increased safety and survey efforts, among other benefits, as described in a previous blog post of mine.

Several study groups around the world have been applying this new technology to a great variety of research applications, aiding in the conservation of certain areas and their respective fauna and flora. Examples of these studies include forest monitoring and tree cover analyses, .

Using drones for forest monitoring and tree cover analyses allows for many applications, such as biodiversity and tree height monitoring, forest classification and inventory, and plant disease and detection. The Ugalla Primate Project, for example, performed an interesting study on tree coverage mapping in western Tanzania (Figure 1).

Figure 1: Tree coverage analyses in Tanzania.
Source: Conservation Drones, 2016.

 

The access to this data (not possible before from the ground) and the acquired knowledge on tree density and structure were important to better understand how wild primates exploit a mosaic landscape. Here is a video about this project:

 

Forest restoration activities can also be monitored by drones. Rainforests around the world have been depleted through deforestation, partly to open up space for agriculture. To meet conservation goals, large areas are being restored to rainforests today (Elsevier 2015). It is important to monitor the success of the forest regeneration and to ensure that the inspected area is being replenished with the right vegetation. Since inspection events can be costly, labor intensive and time consuming, drones can facilitate these procedures, making the monitoring process more feasible.

Zahawi et al. (2015) conducted an interesting study in Costa Rica, being able to keep up with the success of the forest regeneration. They were also able to spot many fruit-eating birds important for forest regeneration (eg. mountain thrush, black guan and sooty-capped bush tanager). Researchers concluded that the automation of the process lead to equally accurate results.

Drones can also be used to inspect areas for illegal logging and habitat destruction. Conservationists have struggled to identify illegal activities, and the use of drones can accelerate the identification process of these activities and help to monitor their spread and ensure that they do not intersect with protected areas.

The Amazon Basin Conservation Association Los Amigos conservancy concession (LACC) has been monitoring 145,000 hectars of the local conservation area. Illegal gold mining and logging activities were identified (Figure 2) and drones have aided in tracking the spread of these activities and the progress of reforestation efforts.

Figure 2: Identification of illegal activities in the Amazon Basin.
Source: NPR, 2015.

 

Another remarkable project was held in Mexico, in one of the most important sites for monarch butterflies in the country: the Monarch Butterfly Biosphere Reserve. Around 10 hectars of vital trees were cut down in the reserve during 2013-2015, and a great decrease of the monarch population was perceived. The reserve did not allow researchers to enter in the area for inspection due to safety concerns. Therefore, drones were used and were able to reveal the illegal logging activity (Figure 3).

Figure 3: Identification of illegal logging at the Monarch Butterfly Biosphere Reserve, Mexico.
Source: Take Part, 2016.

 

Regarding the use of drones for mapping vulnerable areas, this new technology can be used to map potential exposed areas to avoid catastrophes. Concerning responses to fires or other natural disasters, drones can fly immediately, while planes and helicopters require a certain time. The drone material also allows for operating successfully under challenging conditions such as rain, snow and high temperatures, as in the case of fires. Data can be assessed in real time, with no need to have firefighters or other personnel at a dangerous location anymore. Drones can now fulfill this role. Examples of drone applications in this regard are the detection, monitoring and support for catastrophes such as landslides, tsunamis, ship collisions, volcanic eruptions, nuclear accidents, fire scenes, flooding, storms and hurricanes, and rescue of people and wildlife at risk. In addition, the use of a thermal image camera can better assist in rescue operations.

Researchers from the Universidad Politécnica de Madrid (UPM) are developing a system to detect forest fires by using a color index (Cruz et al. 2016). This index is based on vegetation classification techniques that have been adapted to detect different tonalities for flames and smoke (Figure 4). This new technique would result in more cost-effective outcomes than conventional systems (eg. helicopters, satellites) and in reaching inaccessible locations.

Figure 4: Fire detection with Forest Fire Detection Index (FFDI) in different scenes.
Source: UPM, 2016.

 

Marine debris detection by drones is another great functionality. The right localization and the extent of the problem can be detected through drone footage, and action plans for clean-ups can be developed.

A research conducted by the Duke University Marine Lab has been detecting marine debris on beaches around the world. They indicate that marine debris impacts water quality, and harms wildlife (eg. whales, sea birds, seals and sea turtles) that might confuse floating plastic with food. You can read a bit more about their research and its importance for conservation ends here.

Drones are also being extensively used for wildlife monitoring. Through drone footage, researchers around the world have been able to detect and map wildlife and habitat use, estimate densities and evaluate population status, detect rare behaviors, combat poaching, among others. One of the main benefits of using a drone instead of using helicopters or airplanes, or having researchers in the area, is the lower disturbance it may cause on wildlife.

A research team from Monash University is using drones for seabird monitoring in remote islands in northwestern Australia (Figure 5). After some tests, researchers were able to detect which altitude (~75 meters) the drone would not cause any disturbances to the birds. Results achieved by projects like this should be used in the future for approaching the species safely.

Figure 5: Photograph taken by a drone of a crested tern colony on a remote island in Australia.
Source: Conservation Drones, 2014.

 

Drones are also being used to combat elephant and rhino poaching in Africa. They are being implemented to predict, trace, track and catch suspects of poaching. The aim is to reduce the number of animals being killed for the detusking and dehorning practices and the illegal trade. You can read more about this theme here. The drone application on combating one of these illegal practices is also shown here in this video.

As if the innovation of this device alone was not enough, drones are also being used to load other tools. A good example is the collection of whale breath samples by attaching Petri dishes or sterile sponges in the basal part of the drones.

The collection of lung samples allows many health-monitoring applications, such as the analysis of virus and bacteria loads, DNA, hormones, and the detection of environmental toxins in their organisms. This non-invasive physiological tool, known as “Snotbot”, allows sampling collection without approaching closely the individuals and with minimal or no disturbance of the animals. The following video better describes about this amazing project:

It is inspiring to look at all of these wonderful applications of drones in conservation research. Our GEMM Lab team is already applying this great tool in the field and is hoping to support the conservation of wildlife.

 

 

References

Conservation Drones. 2014. Conservation Drones for Seabird Monitoring. Available at: https://conservationdrones.org/2014/05/05/conservation-drones-for-seabird-monitoring/

Conservation Drones. 2016. Tree cover analyses in Tanzania in collaboration with Envirodrone. Available at: https://conservationdrones.org/2016/09/17/tree-cover-analyses-in-tanzania-in-collaboration-with-envirodrone/

Cruz H, Eckert M, Meneses J and Martínez JF. 2016. Efficient Forest Fire Detection Index for Application in Unmanned Aerial Systems (UASs). Sensors 16(893):1-16.

Elsevier. 2015. Drones Could Make Forest Conservation Monitoring Significantly Cheaper: new study published in the Biological Conservation wins Elsevier’s Atlas award for September 2015. Available at: https://www.elsevier.com/about/press-releases/research-and-journals/drones-could-make-forest-conservation-monitoring significantly-cheaper

NPR. 2015. Eyes In The Sky: Foam Drones Keep Watch On Rain Forest Trees. Available at: http://www.npr.org/sections/goatsandsoda/2015/05/19/398765759/eyes-in-the-sky-styrofoam-drones-keep-watch-on-rainforest-trees

Take Part. 2016. Drones Uncover Illegal Logging in Critical Monarch Butterfly Reserve. Available at: http://www.takepart.com/article/2016/06/22/drones-uncover-illegal-logging-monarch-butterfly-habitat

UPM. 2016. New automatic forest fire detection system by using surveillance drones. Available at: http://www.upm.es/internacional/UPM/UPM_Channel/News/dc52fff26abf7510VgnVCM10000009c7648aRCRD

Zahawi RA, Dandois JP, Holl KD, Nadwodny D, Reid JL and Ellis EC. 2015. Using lightweight unmanned aerial vehicles to monitor tropical forest recovery. Biological Conservation 186:287–295.

 

The five senses of fieldwork

By Leila Lemos, PhD student

 

This summer was full of emotions for me: I finally started my first fieldwork season after almost a year of classes and saw my first gray whale (love at first sight!).

During the fieldwork we use a small research vessel (we call it “Red Rocket”) along the Oregon coast to collect data for my PhD project. We are collecting gray whale fecal samples to analyze hormone variations; acoustic data to assess ambient noise changes at different locations and also variations before, during and after events like the “Halibut opener”; GoPro recordings to evaluate prey availability; photographs in order to identify each individual whale and assess body and skin condition; and video recordings through UAS (aka “drone”) flights, so we can measure the whales and classify them as skinny/fat, calf/juvenile/adult and pregnant/non-pregnant.

However, in order to collect all of these data, we need to first find the whales. This is when we use our first sense: vision. We are always looking at the horizon searching for a blow to come up and once we see it, we safely approach the animal and start watching the individual’s behavior and taking photographs.

If the animal is surfacing regularly to allow a successful drone overflight, we stay with the whale and launch the UAS in order to collect photogrammetry and behavior data.

Each team member performs different functions on the boat, as seen in the figure below.

Figure 1: UAS image showing each team members’ functions in the boat at the moment just after the UAS launch.
Figure 1: UAS image showing each team members’ functions in the boat at the moment just after the UAS launch.

 

While one member pilots the boat, another operates the UAS. Another team member is responsible for taking photos of the whales so we can match individuals with the UAS videos. And the last team member puts the calibration board of known length in the water, so that we can later calculate the exact size of each pixel at various UAS altitudes, which allows us to accurately measure whale lengths. Team members also alternate between these and other functions.

Sometimes we put the UAS in the air and no whales are at the surface, or we can’t find any. These animals only stay at the surface for a short period of time, so working with whales can be really challenging. UAS batteries only last for 15-20 minutes and we need to make the most of that time as we can. All of the members need to help the UAS pilot in finding whales, and that is when, besides vision, we need to use hearing too. The sound of the whale’s respiration (blow) can be very loud, especially when whales are closer. Once we find the whale, we give the location to the UAS pilot: “whale at 2 o’clock at 30 meters from the boat!” and the pilot finds the whale for an overflight.

The opposite – too many whales around – can also happen. While we are observing one individual or searching for it in one direction, we may hear a blow from another whale right behind us, and that’s the signal for us to look for other individuals too.

But now you might be asking yourself: “ok, I agree with vision and hearing, but what about the other three senses? Smell? Taste? Touch?” Believe it or not, this happens. Sometimes whales surface pretty close to the boat and blow. If the wind is in our direction – ARGHHHH – we smell it and even taste it (after the first time you learn to close your mouth!). Not a smell I recommend.

Fecal samples are responsible for the 5th sense: touch!

Once we identify that the whale pooped, we approach the fecal plume in order to collect as much fecal matter as possible (Fig.2).

Figure 2: A: the poop is identified; B: the boat approaches the feces that are floating at the surface (~30 seconds); C: one of the team members remains at the bow of the boat to indicate where the feces are; D: another team member collects it with a fine-mesh net. Filmed under NOAA/NMFS permit #16111 to John Calambokidis).
Figure 2: A: the poop is identified; B: the boat approaches the feces that are floating at the surface (~30 seconds); C: one of the team members remains at the bow of the boat to indicate where the feces are; D: another team member collects it with a fine-mesh net. Filmed under NOAA/NMFS permit #16111 to John Calambokidis).

 

After collecting the poop we transfer all of it from the net to a small jar that we then keep cool in an ice chest until we arrive back at the lab and put it in the freezer. So, how do we transfer the poop to the jar? By touching it! We put the jar inside the net and transfer each poop spot to the jar with the help of water pressure from a squeeze bottle full of ambient salt water.

Figure 3: Two gray whale individuals swimming around kelp forests. Filmed under NOAA/NMFS permit #16111 to John Calambokidis).
Figure 3: Two gray whale individuals swimming around kelp forests. Filmed under NOAA/NMFS permit #16111 to John Calambokidis).

 

That’s how we use our senses to study the whales, and we also use an underwater sensory system (a GoPro) to see what the whales were feeding on.

GoPro video of mysid swarms that we recorded near feeding gray whales in Port Orford in August 2016:

Our fieldwork is wrapping up this week, and I can already say that it has been a success. The challenging Oregon weather allowed us to work on 25 days: 6 days in Port Orford and 19 days in the Newport and Depoe Bay region, totaling 141 hours and 50 minutes of effort. We saw 195 whales during 97 different sightings and collected 49 fecal samples. We also performed 67 UAS flights, 34 drifter deployments (to collect acoustic data), and 34 GoPro deployments.

It is incredible to see how much data we obtained! Now starts the second part of the challenge: how to put all of this data together and find the results. My next steps are:

– photo-identification analysis;

– body and skin condition scoring of individuals;

– photogrammetry analysis;

– analysis of the GoPro videos to characterize prey;

– hormone analysis laboratory training in November at the Seattle Aquarium

 

For now, enjoy some pictures and a video we collected during the fieldwork this summer. It was hard to choose my favorite pictures from 11,061 photos and a video from 13 hours and 29 minutes of recording, but I finally did! Enjoy!

Figure 4: Gray whale breaching in Port Orford on August 27th. (Photo by Leila Lemos; Taken under NOAA/NMFS permit #16111 to John Calambokidis).
Figure 4: Gray whale breaching in Port Orford on August 27th. (Photo by Leila Lemos; Taken under NOAA/NMFS permit #16111 to John Calambokidis).

 

Figure 5: Rainbow formation through sunlight refraction on the water droplets of a gray whale individual's blow in Newport on September 15th. (Photo by Leila Lemos; Taken under NOAA/NMFS permit #16111 to John Calambokidis).
Figure 5: Rainbow formation through sunlight refraction on the water droplets of a gray whale individual’s blow in Newport on September 15th. (Photo by Leila Lemos; Taken under NOAA/NMFS permit #16111 to John Calambokidis).

 

Likely gray whale nursing behavior (Taken under NOAA/NMFS permit #16111 to John Calambokidis):

Unmanned Aircraft Systems: keep your distance from wildlife!

By Leila Lemos, Ph.D. Student, Department of Fisheries and Wildlife, OSU

Unmanned aircraft systems (UAS) or “drones” are becoming commonly used to observe natural landscapes and wildlife. These systems can provide important information regarding habitat conditions, distribution and abundance of populations, and health, fitness and behavior of the individuals (Goebel et al. 2015, Durban et al. 2016).

The benefits for the use of UAS by researchers and wildlife managers are varied and include reduced errors of population estimations, reduced observer fatigue, increased observer safety, increased survey effort, and access to remote settings and harsh environments (Koski et al. 2010, Vermeulen et al. 2013, Goebel et al. 2015, Smith et al. 2016). Importantly, data gathered from UAS can provide needed information for the conservation and management of several species. Although it is often assumed that wildlife incur minimal disturbance from UAS due to the reduced noise compared to traditional aircraft used for wildlife monitoring (Acevedo-Whitehouse et al. 2010), the impacts of UAS on most wildlife populations is currently unexplored.

Several studies have tried to comprehend the effects of UAS flights over animals and so far there is no evidence of behavioral disturbance. For instance Vermeulen et al. (2013) conducted a study where authors observed a group of elephants’ reaction or warning behavior while a UAS passed ten times over the individuals at altitudes of 100 and 300 meters, and no disturbance was recorded. Furthermore, a study conducted by Acevedo-Whitehouse et al. (2010) reported that six different species of large cetaceans (Bryde’s whale, fin whale, sperm whale, humpback whale, blue whale and gray whale) did not display avoidance behavior when approached by the UAS for blow sampling, suggesting that the system caused minimal distress (negative stress) to the individuals.

However, the fact that we cannot visually see an effect in the animal does not mean that a stress response is not occurring. A study analyzed the effects of UAS flights on movements and heart rate responses of American black bears in northwestern Minnesota (Ditmer et al. 2015). It was observed that all bears, including an individual that was hibernating, responded to UAS flights with increased heart rates (123 beats per minute above the pre-flight baseline). In contrast, no behavioral response by the bears was recorded (Figure 1).

By Leila Lemos, Ph.D. Student, Department of Fisheries and Wildlife, OSU Unmanned aircraft systems (UAS) or “drones” are becoming commonly used to observe natural landscapes and wildlife. These systems can provide important information regarding habitat conditions, distribution and abundance of populations, and health, fitness and behavior of the individuals (Goebel et al. 2015, Durban et al. 2016). The benefits for the use of UAS by researchers and wildlife managers are varied and include reduced errors of population estimations, reduced observer fatigue, increased observer safety, increased survey effort, and access to remote settings and harsh environments (Koski et al. 2010, Vermeulen et al. 2013, Goebel et al. 2015, Smith et al. 2016). Importantly, data gathered from UAS can provide needed information for the conservation and management of several species. Although it is often assumed that wildlife incur minimal disturbance from UAS due to the reduced noise compared to traditional aircraft used for wildlife monitoring (Acevedo-Whitehouse et al. 2010), the impacts of UAS on most wildlife populations is currently unexplored. Several studies have tried to comprehend the effects of UAS flights over animals and so far there is no evidence of behavioral disturbance. For instance Vermeulen et al. (2013) conducted a study where authors observed a group of elephants’ reaction or warning behavior while a UAS passed ten times over the individuals at altitudes of 100 and 300 meters, and no disturbance was recorded. Furthermore, a study conducted by Acevedo-Whitehouse et al. (2010) reported that six different species of large cetaceans (Bryde’s whale, fin whale, sperm whale, humpback whale, blue whale and gray whale) did not display avoidance behavior when approached by the UAS for blow sampling, suggesting that the system caused minimal distress (negative stress) to the individuals. However, the fact that we cannot visually see an effect in the animal does not mean that a stress response is not occurring. A study analyzed the effects of UAS flights on movements and heart rate responses of American black bears in northwestern Minnesota (Ditmer et al. 2015). It was observed that all bears, including an individual that was hibernating, responded to UAS flights with increased heart rates (123 beats per minute above the pre-flight baseline). In contrast, no behavioral response by the bears was recorded (Figure 1).
Figure 1: (A) Movement rates (meters per hour) of an adult female black bear with cubs prior to, during, and after a UAS flight (gray bar); (B) The corresponding heart rate (beats per minute) of the adult female black bear. Source: Modified from Figure 1 from Ditmer et al. 2015.

 

Therefore, behavioral analysis alone may not be able to describe the complete effects of UAS on wildlife, and it is important to consider other possible stress responses of wildlife.

Regarding marine mammals, only a few studies have systematically documented the effects of UAS on these animals. A review of these studies was produced by Smith et al. (2016) and the main factors influencing behavioral disturbance were identified as (1) noise and visual stimulus (from the UAS or its shadow), and (2) flight altitude of the UAS. Thus, studies that approach marine mammals closely with UAS (e.g., blow sampling in cetaceans) should be closely monitored for behavioral reactions because the noise level and visual stimulus will likely be increased.

Fortunately, when UAS work is applied to cetaceans and sirenians (manatees and dugongs) the air-water interface acts as a barrier to sound so these animals are unlikely to be acoustically disturbed by UAS. However, acoustic detection and response are still possible when an animal’s ears are exposed in the air during a surfacing event.

The best way to minimize stress responses in wildlife is to use caution while operating UAS at any altitude. According to National Oceanic and Atmospheric Administration (NOAA), “UAS can also be disruptive to both people and animals if not used safely, appropriately, or responsibly”. Therefore, since 2012, the Federal Aviation Administration (FAA) has required UAS operators in the United States to have a certified and registered aircraft, a licensed pilot, and operational approval, known as Section 333 Exemption (Note: in late August 2016, the 333 will be replaced by a revision to part 107). These authorizations require an air worthiness statement or certificate and registered aircraft. Public entities, like Oregon State University, operate under a certificate of authorization (COA.) As a public entity OSU certifies its own aircraft and sets standards for UAS operators. These permit requirements discourage illegal operations and improves safety.

Regarding marine mammals, all UAS operators should also be aware of The Marine Mammal Protection Act (MMPA) of 1972. This law makes it illegal to harass marine mammals in the wild, which may cause disruption to behavioral patterns, including, but not limited to, migration, breathing, nursing, breeding, feeding, or sheltering. A close UAS approach has the potential to cause harassments to marine mammals, thus federal guidelines recommend keeping a safe distance from these animals in the wild. The required vertical distance is 1000 ft for most marine mammals, but increases for endangered animals such as the North Atlantic right whales with a required buffer of 1500 ft (http://www.nmfs.noaa.gov/pr/uas.html). Therefore, NOAA evaluates all scientific research that use UAS within 1000 ft of marine mammals in order to ensure that the benefits outweigh possible hazards. NOAA distributes research permits accordingly.

Of course, with new technology the rules are always changing. In fact, last week the Department of Transportation (DOT) and the FAA finalized the first operational rules for routine commercial use of small UAS. These new guidelines aim to support new innovations in order to spur job growth, advance critical scientific research and save lives, and are designed to minimize risks to other aircraft and people and property on the ground. These new regulations include several requirements (e.g., height and speed restrictions) and hopefully allow for a streamlined system that enables beneficial and exciting wildlife research.

For my PhD project we are using UAS to collect aerial images from gray whales in order to describe behavioral patterns and apply a photogrammetry methodology. Through these methods we will determine the overall body condition and health of the individuals for comparison to variable ambient ocean noise levels. This project is conducted in collaboration with the NOAA Pacific Marine Environmental Lab.

Since October 2015, we have conducted 31 over-flights of gray whales using our UAS (DJI Phantom 3) and no behavioral disturbance has been observed. When over the whale(s) we generally fly between 25 and 40 m above the animals. We have a FAA certified UAS operator and fly under our NOAA/NMFS permit 16111. Prior to each flight we ensure that the weather conditions are safe, the whales are behaving normally, and that no on-lookers from shore or other boats will be disturbed.

Here is a video showing the launch and retrieval of the UAS system, our research vessel, the surrounding Oregon coastline beauty and gray whale individuals. The video includes some interesting footage of a gray whale foraging over a shallow reef, indicating that this UAS flight did not disturb the animal’s natural behavior patterns.

We all have the responsibility to help keep wildlife safe. Here in the GEMM Lab, we commit to using UAS safely and responsibly, and aim to use this new and exciting technology to continue our efforts to better protect and understand marine mammals.

 

References

Acevedo‐Whitehouse K, Rocha‐Gosselin A and Gendron D. 2010. A novel non‐invasive tool for disease surveillance of free‐ranging whales and its relevance to conservation programs. Anim. Conserv. 13(2):217–225.

Ditmer MA, Vincent JB, Werden LK, Tanner JC, Laske TG, Iaizzo PA, Garshelis DL and Fieberg JR. 2015. Bears Show a Physiological but Limited Behavioral Response to Unmanned Aerial Vehicles. Current Biology 25:2278–2283.

Durban JW, Moore MJ, Chiang G, Hickmott LS, Bocconcelli A, Howes G, Bahamonde PA, Perryman WL and Leroi DJ. 2016. Photogrammetry of blue whales with an unmanned hexacopter. Marine Mammal Science. DOI: 10.1111/mms.12328.

Goebel ME, Perryman WL, Hinke JT, Krause DJ, Hann NA, Gardner S and LeRoi DJ. 2015. A small unmanned aerial system for estimating abundance and size of Antarctic predators. Polar Biol. 38(5):619-630.

Koski WR, Abgrall P and Yazvenko SB. 2010. An inventory and evaluation of unmanned aerial systems for offshore surveys of marine mammals. J. Cetacean Res. Manag. 11(3):239–247.

NOAA. Unmanned Aircraft Systems: Responsible Use to Help Protect Marine Mammals. In: http://www.nmfs.noaa.gov/pr/uas.html. Accessed in: 06/12/2016.

Smith CE, Sykora-Bodie ST, Bloodworth B, Pack SM, Spradlin TR and LeBoeuf NR. 2016. Assessment of known impacts of unmanned aerial systems (UAS) on marine mammals: data gaps and recommendations for researchers in the United States1 J. Unmanned Veh. Syst. 4:1–14.

Vermeulen C, Lejeune P, Lisein J, Sawadogo P and Bouché P. 2013. Unmanned aerial survey of elephants. PLoS One. 8(2):e54700.