African elephants adjust speed in response to surface-water constraint on foraging during the dry-season

Abstract

Most organisms need to acquire various resources to survive and reproduce. Individuals should adjust their behavior to make optimal use of the landscape and limit the costs of trade-offs emerging from the use of these resources. Here we study how African elephants Loxodonta africana travel to foraging places between regular visits to waterholes. Elephant herds were tracked using GPS collars during two consecutive dry seasons in Hwange National Park, Zimbabwe. We segmented each individual movement track at each visit to water to define foraging trips, and then used trip-level statistics to build an understanding of movement strategies. Travel speed within these individually-consistent movement bouts was also analyzed to understand if speed was better linked to distance to water or progression in the trip over time. We found that elephants went further from water when drinking less often, which could result from a trade-off between drinking and foraging in less depleted, far from water, places. Speed increased towards the beginning and the end of the trips, and was also greater than observed during the wet season, suggesting that elephants were trying to save time. Numerous short trips traveled at greater speed, particularly when commuting to a different waterhole, was tentatively explained by the inability to drink at specific waterholes due to intra-specific interference. Unexpectedly elephants did not always minimize travel time by drinking at the closest waterhole, but the extra distance traveled remained never more than a few kilometers. Our results show how individuals may adjust movement behavior to deal with resource trade-offs at the landscape scale. We also highlight how behavioral context, here progression in the trip, may be more important than spatial context, here distance to water, in explaining animal movement patterns.

Figure 1. Characteristics of trips conducted between visits to waterholes.

(A) Frequency of trip durations; (B) Relationship between distance traveled and trip duration; (C) Relationship between the maximum distance to waterholes visited at the start/end of the trip and trip duration; (D) Relationship between mean travel speed and trip duration. Note that in the wet season when elephants are not constrained by water availability mean speed is 0.33±0.52 s.d. km/h. All panels show differences between commuting trips (when different waterholes are visited at the beginning and at the end of the trip; white-filled symbols, dotted lines) and looping trips (when the same waterhole is visited at the beginning and at the end of the trip; black-filled symbols, solid lines). Note that the difference between commuting and looping trips in panel (C) is not significant (see Table 1).

Figure 2. Relationship between speed and progression in the trip.

(A) Speed of elephants within trips, in relation to the progression in the trip (expressed as percent of the total duration of the trip). Each line represents data for a single trip. Trips are defined as the movement between visits to waterholes, and elephants are therefore closer to water at the beginning and at the end of the trip. The waterhole at the beginning of the trip is either different (commuting trip) or the same (looping trip) than the waterhole at the end of the trip. A statistical model revealed that speed was best related to progression in the trip, trip type (commuting vs. looping) and some interactions with trip duration (see Table 2). Panel (B) shows the model predictions. Note that in the wet season when elephants are not constrained by water availability mean speed is 0.33±0.52 s.d. km/h.