This study evaluated potential disturbance effects caused to Hawaiian monk seals and birds in the Northwestern Hawaiian Islands during flights with an APH-22 hexacopter.

Unmanned Aerial Systems (UAS) have great potential for use in Hawaiian monk seal conservation research, as a non-invasive tool that allows researchers to easily survey hard-to-reach areas.

Possible applications of UAS technology include population count surveys and body condition measurements using photogrammetry.

We investigated whether APH-22 flights disturb target and non-target study species, Hawaiian monk seals and birds.

Data collection:

Behavioral disturbance data was collected during flights of a remotely operated APH-22 hexacopter in the Northwestern Hawaiian Islands during September 2015 and April and May 2018.

APH-22 flights:

• Hand-launched from small boat or shore within line of sight
• Pilot in command, ground station operator, visual observer
• 3 altitudes: approximately 50 m, 30 m, and 10 m

Seal behaviors recorded initially and at each altitude

Bird behaviors recorded during 4 flight periods: before flight, during launch, during flight, and during landing

Data analysis:

Seal behavioral responses compared by altitude, location, age, and initial behavior

Bird behavioral responses compared by group size, flight period, and species

3 categories of seal behavior: No change, minor behavioral changes, position changes

2 categories of bird behavior: Flushed/did not flush, approached UAS/did not approach

Descriptive plots used to compare seal and bird behavioral responses between groups

Generalized linear mixed-effects model of seal behavior as a function of altitude, location, age, and initial behavior

Model chosen by backwards selection based on AIC score and significance level of individual variables

Seals displaying a variety of observed behaviors. Clockwise from bottom left: No change, raised head, directed gaze.

Behavioral response categories:

• No change
• Minor behavioral changes: Raised head, glanced at UAS, directed gaze
• Position changes: Changed body position, moved away
   

Behavioral response observed in Hawaiian monk seals during hexacopter UAS flights

Coefficients of best generalized linear mixed-effects model for three behavioral categories. Green boxes indicate variables that were included in the selected model.

Behavioral Response Categories:

• Flushed: Took off from beach/vegetation
  Did not flush: Could include no response, looked at UAS, approached UAS, other
• Approached: Swooped towards UAS or followed UAS in flight
  Did not approach: Could include no response, looked at UAS, flushed, other

Behavioral response observed in birds during hexacopter UAS flights

Seals:

• Seals change behaviors more at lower altitudes. Most seals (81%) did not change behaviors above 25 m altitude
• Raised head was most common behavioral change in seals, especially below 14 m (68%).
• Directed gaze, changes in body position and movement away were rare (0.5% of all observations).
• Behavioral changes less frequent at French Frigate Shoals

Birds:

• Larger groups flushed more and smaller groups or individuals approached more
• Most flushed during launch (46%) or during flight (43%). Most approached during flight (74%).
• Disturbance varied by species.
  Noddies and boobies flushed most
  Noddies, boobies, terns, and frigates approached most
• Most birds that flushed returned within 2 minutes
• No birds made contact with UAS

Disturbance caused to seals and birds by APH-22 hexacopter flights was minimal, making APH-22 an appropriate tool for future surveys of these species in the Northwestern Hawaiian Islands. Flying with higher resolution cameras will allow researchers to pilot the UAS at higher altitudes and collect the same quality of data with less disturbance.

The results of this study and the capabilities of UAS for studying Hawaiian monk seals remotely and in hard-to-reach locations indicate that UAS are a promising tool for future conservation research.

Acknowledgements:

We would like to thank the teams from the Hawaiian Monk Seal Research Program that conducted UAS flights and collected behavioral data in the Northwestern Hawaiian Islands. Special thanks also to Stacie Robinson and Jessica Bohlander for their guidance and mentorship in this research. Research was conducted under MMPA/ESA permit #16632-00 & 16632-02, PMNM-2015-01 & PNMN-2018-01. This research is supported by the NOAA Hollings Undergraduate Scholarship Program.