Skip to main content

Table 1 Experimental protocol for each species

From: How often should dead-reckoned animal movement paths be corrected for drift?

Animal (number (n) assessed)

Scheduled GPS frequency

DR frequency

Approx. DR length

Extrapolated speeds

Capture and deployment method

Lion

(n = 10)

1 Hz

1 Hz

2 weeks

DBA-based speed

Due to high variability in step gaits, m- and c-values were computed per individual from VeDBA ~ GPS-derived speed regression (cf. [23])

Prides were lured to bait using audio recordings and individuals were anaesthetised at night according to SANParks operational procedures—detailed in SANParks’ ‘Standard Operating Procedures for the Capture, Transportation and Maintenance in Holding Facilities of Wildlife’

Units were mounted to a Litetrack collar [https://www.lotek.com]. Collars were loose enough to allow three fingers to pass through

Penguin

on land (walking)

(n = 15 –

2 DR paths per individual (out- and inbound)

1 Hz

10 Hz

30 min

DBA-based speed

Due to having a constant step gait, m- values were selected (c = 0) per individual based on the best scaling relative to GPS path pre-VP correction

Penguins were caught at the nest during the chick rearing season using the clipboard method [74] and cormorants were caught at the nest during the chick rearing season via a crook on the end of a long pole (cf. [75]). Birds were blind folded and restrained on a researcher's knees

Devices fitted longitudinally to the base of the spine using Tesa® tape [74, 76, 77]

Penguin

at sea (diving)

(n = 15)

1 Hz

2 Hz

1.5 days

Change between constant values (according to behaviour-type) and vertical movement-based speed

speed = 0.416 m/s (cf. [78]) when depth ≤ 0.3 m (cf. [79])

speed = 2.1 m/s (cf. [80, 81]) when depth > 0.3 m and absolute values of pitch were < 10o

speed = Δd/tan(θ • π/180) (upper cap of speed derived this way = 3 m/s) when depth > 0.3 m and absolute values of pitch were ≥ 10°

Cormorant at sea (flying and diving)

(n = 15)

1 Hz

10 Hz

8 h

Change between constant values (according to behaviour-type) and vertical movement-based speed

speed = 12 m/s when flying (derived from the heave acceleration (cf. [69])

speed = 0.1 m/s when resting at the sea surface (derived from depth sensor and lack of dynamic acceleration)

speed = Δd/tan(θ • π/180) (upper cap of speed derived this way = 3 m/s) during the ascents and descents of dives

speed = 0.4 m/s during the bottom phase of dives

Tropicbird at sea (flying)

(n = 7)

1 fix every minute

10 Hz

3 h

GPS-based speed

speed = Haversine distance between GPS fixes divided by the time period between values and linearly interpolated (cf. [82]). Speed values overwritten as 0.1 m/s when birds were resting at sea surface

Devices were placed in a zip-lock bag, inside unheated heat shrink wrap and fixed longitudinally to the back feathers [72] using Tesa® tape [76]

  1. ‘Δd/tan(θ· π/180)’ refers to the rate change of depth (m/s) divided by the tangent of the body pitch (converted from degrees to radians). DR =  dead-reckoning and DBA = dynamic body acceleration. m- and c-values represent the (multiplicative) coefficient (gradient) and constant (intercept) of the VeDBA–speed regression [speed = VeDBA · m + c]