Divergent Circadian Foraging Strategies in Response to Diurnal Predation Versus Persistent Rain in Asian Weaver Ant,Oecophylla smaragdina,Suggest Possible Energetic Trade-offs

Abstract

The study of chronobiology of foraging behavior in social insects offers valuable models for the investigation of circadian rhythms. We scored hourly nest entries and exits of Oecophylla smaragdina (Asian weaver ant) workers in 9 active non-polydomous nests on days with and without rain and with and without a primarily diurnal predator present. After determining that Oecophylla display a high nest fidelity, we focused exclusively on analyzing nest entry counts: we found a significant decrease in overall entry counts of individual ants on rainy days compared with non-rainy days (p < 0.0001). They usually maintain a typical diurnal pattern of foraging activity; however, that regularity was often distorted during rainy periods but appeared to quickly revert to typical patterns following rain. This lack of compensatory foraging activity following a period of rain supports the hypothesis that these ants have enough food reserves to withstand a pure masking-induced suppression of foraging activity. Predation through bird anting, too, decreased foraging activity but appeared to cause a reversal in foraging activity timing from diurnal to nocturnal foraging. Daily periodicity of foraging was significantly disrupted in most nests during rain; however, daily foraging periodicity was disrupted in only one nest due to presence of predators. Thus, rain and predation both exert significant impacts on the overall foraging activity of Asian weaver ants, but while persistent pressure from rain seemed to primarily cause masking (diminution) of circadian foraging activity, predation restricted to the daytime resulted in phase-inversion to nocturnal foraging activity, with little diminution. This is consistent with different energetic strategies being used in response to different pressures by this species.

Keywords

Asian weaver ants activity rhythm fidelity foraging plasticity masking temporal niche plasticity

Regular behavioral events in ant societies involve foraging for food (Fujioka et al., 2019; Pinter-Wollman et al., 2012) and nest defense (Fujioka et al., 2019; Hölldobler and Wilson, 1990)—both of which either enhance or decline with seasons (El Keroumi et al., 2022) or at a specific time of the day (Villalta et al., 2020). Whatever the pattern, the behavioral competencies are often synchronized to environmental situations, particularly with climatic functioning and stability (Button et al., 2020), guided by the organism’s internal clock (Das and Gordon, 2023). The foraging performance or behavioral cycles in ants are often shaped by seasonal or daily climatic shifts due to their poikilothermic physiological limitations to work with thermal constraints (Miyazaki et al., 2014). Ants engage in collaborative foraging activities that require individual efforts but are coordinated through group behavior (Reeves and Moreau, 2019). However, the discipline of foraging socioecology encounters major challenges in decoding the collective behavior of a system as potentially vast and complex as an ant colony (Hölldobler and Wilson, 1990; Traniello, 1989). Ant workers often exit and enter the nest during foraging throughout day and night in shifts and at different intensities. Although ant nests usually have many active outlets (Pinter-Wollman et al., 2012), they are likely to reside in a communal setting and exhibit spatial foraging fidelity (Crall et al., 2018), so workers are more likely to drive some nestmates to commence foraging by biasing access to spatially localized nest information (Evans et al., 2021). It has been shown that workers exhibit remarkable repeatability in their spatial fidelity patterns across individual days, with workers returning to the same spatial zones within the nest in predictable patterns on successive days (Crall et al., 2018; Hanisch et al., 2023). As a result, foraging individuals tend to utilize the same entry across time (Cane et al., 1996; Lehue and Detrain, 2020; Mersch et al., 2013).

The primary concern is understanding how the organism’s internal clock is optimized to express and regulate itself in a specific natural environment (Kronfeld-Schor et al., 2013). Behavioral rhythms are thought to be a combination of endogenous components caused by the organism’s endogenous oscillators, and external components which can alter phase of the oscillators themselves, or without altering their phase, may nevertheless “mask” the expression of the oscillators’ normal output (Rietveld et al., 1993). The phenomenon of “masking” enables an organism to efficiently and promptly adjust to alterations in its surroundings when waiting for circadian realignment might be too slow or too costly (Rietveld et al., 1993). Insects commonly exhibit “masking” in their diurnal patterns, which can ultimately alter their daily behavioral routine (Page, 1989). Ants also exhibit variations in their regular activities based on the time of day, influenced by diverse exogenous regulators (Roces and Nunez, 1996). Therefore, their capacity to adjust to environmental change is crucial in regulating their behavioral rhythm and performance (Jürgen Stelzer et al., 2010). Masking of diurnal foraging behavior has been seen in response to heat in harvester ants, Pogonomyrmex sp. (Friedman, 2019), and in response to many causes in carpenter ants (Camponotus sp.). For example, the diurnal clock of Camponotus compressus is highly flexible and often masking to the light-dark cycles, and it often loses after mating (Lone, 2023; Sharma, Lone, and Goel, 2004; Sharma, Lone, Goel, and Chandrashekaran, 2004; Sharma, Lone, Mathew, et al., 2004). Likewise, Camponotus rufipes exhibits significant plasticity in its feeding schedule within a social setting (Mildner and Roces, 2017).

In addition to thermal stress and photosensitivity, other exogenous factors, such as rain, also impact foraging incentives among the ants. For example, the African weaver ant, Oecophylla longinoda, exhibits foraging tendencies predominantly during sunny days in summer, but in rain, they become nocturnal, and their foraging activity during the day decreases remarkably (Vayssières et al., 2011). Like weaver ants, the leaf-cutting Atta sp. may also stop their foraging behavior during rain (Farji-Brener et al., 2018). Though Atta are predominantly nocturnal in the dry summer, they become diurnal when exposed to rain at night (Rockwood, 1975) and may follow similar foraging practices throughout the rainy season as in the summer (Farji-Brener et al., 2018). Such occurrences indicate that rain disrupts their regular foraging activity, but they resume their normal foraging activity cycle following the rainfall (Hodgson, 1955).

Since circadian rhythms may provide an oscillator (the performing animal) with a survival advantage and a means of evading danger (predation avoidance), they significantly shape and maintain the population structure (Brown and Braithwaite, 2005). Birds, particularly those from the passerine family, are well known for preying on a variety of insects, including ants (Philpott et al., 2005). In the presence of predators, some Neotropical army ant species, like Eciton sp. and Neivamyrmex sp., can show diurnal to nocturnal foraging preferences. As a result of parasitic fly interventions, the feeding cycle of Atta cephalotes is delayed throughout the day and induces nocturnal activities (Orr, 1992). The diel specialization phenomenon enables species to coexist in the same habitat, particularly in cases where season-based phenological changes or temporal separation of foraging times are moderated (O’Donnell et al., 2021).

Considering the above, we proposed that exogenous stimulators, such as precipitation and predation pressure, may have a significant but different masking role within a single species of ant’s everyday activities and periodic behaviors. On the assumption that ants can maintain their foraging activity schedule in the face of exogenous “masking” factors like rain and predation pressure, we investigated how quickly they reverted to their regular schedule after the mediators were withdrawn.

Materials And Methods

Focal Species

Oecophylla smaragdina (Formicidae: Hymenoptera) is renowned for its skill in building arboreal nests. These nests are commonly found throughout the temperate Australasia region and typically consist of polydomous colonies. These colonies are characterized by multiple leaf nests intricately woven together using silk produced by their larvae. The dimensions of their nest may vary from a solitary leaf to several leaves and can house a few individuals to a maximum of 500,000 ants (Ambika and Nalini, 2019). They have a distinct caste system, including workers, queens, and males who engage in different social activities (Hölldobler and Wilson, 1990). The major workers of Oecophylla are accountable for the nest construction, foraging for procurement of food, and safeguarding of the territory from both conspecific and heterospecific intruders, whereas the minor workers partake in the responsibility of nurturing the offspring. The workers may engage in foraging activities at a considerable distance from the source of sustenance and frequently carry out such tasks collectively. They exhibit a notable degree of fidelity in their nest-returning behavior and engage in a division of labor using distinct groups working in shifts. The workers can tolerate temperatures ranging from a minimum of 15 °C to a maximum of 35 °C, and such adaptability enables them to inhabit various tropical climatic zones spanning from Southeast Asia to Australia (Chua et al., 2022). Oecophylla is classified as omnivores as they consume a variety of food sources, including nectar, fruits, and animal flesh (Crozier et al., 2010; Vidhu and Evans, 2015). Their lifespan is estimated to be approximately 3 years in their natural habitat (Mukhopadhyay and Sannigrahi, 1993), and colony lifespan is approximately 10 years.

Study Site and Nest Selection

To observe the foraging activity patterns of O. smaragdina, 9 wild leaf nests (nests 1-9) were chosen from different locations in Midnapore, West Bengal, India (22.3660°N, 87.5503°E). The nests were from different mango trees (Mangifera indica) separated by approximately 200 m to confirm their separate colonies (non-polydomous nature). The selected nests were relatively uniform in size, with an average diameter of 20 cm, built of around 16 to 26 mango leaves, and positioned at a height of 3 m above the ground, allowing video cameras to be easily fixed in front of them for data collection. The data collection involved systematic tracking (by manually observing the video data) and recording the periodicity of ant movement as they entered and exited the nests via the primary nest outlet on a 24-h day scale. The 9 selected nests were divided into 3 groups of 3 nests each (group 1: nests 1-3, group 2: nests 4-6, and group 3: nests 7-9) based on the circumstances in which ant traffic (entry and exit) was documented at each hour of a day for 3 different situations (named as episodes). Ants’ entry-exit data for nests 1 to 3 (nest group 1) were collected in the summer (dry season) (March-June; mean temperature: 31-38 °C, relative humidity: 70%-80%, average rainfall: 100-120 mm); ant activities for nests 4-6 (nest group 2) were collected during the rainy season (July-October; mean temperature: 28-32 °C, relative humidity: 95%-100%, average rainfall: 300-350 mm). Furthermore, data related to nests 7-9 (nest group 3) were collected during the specified no rain days characterized by the presence a of jungle babbler bird, Argya striata (Leiothrichidae: Passeriformes) and Indian myna, Acridotheres tristis (Sturnidae: Passeriformes), that exhibited high levels of activity in hunting ants by positioning in close proximity to their nests (“predation”) (March-July; mean temperature: 31-38 °C, relative humidity: 70%-80%, average rainfall: 30-80 mm).

Only 30-day data for a single nest of a nest group for a certain incident (dry season, rainy season, or predation) were collected. For instance, only 30 days of “rain” and “no rain” day data were collected for each of nests 1, 2, and 3 in the summer days (out of possibly 122 days from March through June). The 30 data collection days varied from nest to nest (1-3), but the unifying criterion was that there were maximum dry days on those data-gathering days. Similarly, 30 days of data were collected for the rainy season (wet days) from a single nest (nest 4, 5, or 6). For example, during the rainy season (July-October), only 30 days of rainy day data were recorded for each of the nests 4, 5, and 6. Similar to before, the 30 days may vary from nest to nest (4-6); however, the days on which data were gathered had heavy rainfall (it rained for 8-18 h a day). Finally, only 30-day data for predation (bird anting) were gathered from a single nest (nests 7, 8, and 9). For example, 30-day data confronted to bird’s predation pressure were gathered each for nests 7, 8, and 9. Since the birds’ anting may differ from nest to nest depending on the availability of their other food and their breeding season (March-April and July-September months), the data collection days may also vary among the nests (7-9); however, all of the days on which data were gathered had bird’s predatory activity to the ants. Oecophylla’s entry-exit activity in response to increased predation pressure in the presence of the jungle babbler and Indian myna bird could change since the bird frequently captures and consumes the ant. According to our video recordings, the passerine exhibits a regular tendency to visit the Oecophylla nest approximately 6-7 times per day (for a period of 30-60 min per turn) to procure the victims upon their nest emergence (Supplemental Video 1: https://drive.google.com/file/d/1l6lyEfpnAcyHMfGiB90Ndgq7e_tvGKJ9/view?usp=sharing), Supplemental Figure 1. The predation rate of the bird was notably higher during daytime than during the night hours.

Data Collection

Although an Oecophylla nest typically has many entry-exit points, a primary port is normally assigned for their everyday transit. Prior to recording foraging moves, the primary entry-exit port was identified through their mass movements (preliminary observation during foraging time). The experimental camera setup was designed specifically to monitor and capture the movement of ants at the primary entry and exit point of the nest (Supplemental Video 1: https://drive.google.com/file/d/1l6lyEfpnAcyHMfGiB90Ndgq7e_tvGKJ9/view?usp=sharing), Supplemental Figure 1. A CCTV surveillance camera (HIKVISION, 5 MP Ultra HD Outdoor Bullet Infrared) was strategically positioned to focus on a particular area, enabling clear visualization of the incoming and outgoing ant activity. The camera boasts 5 MP resolution at 20 frames per second (fps), features water and dust resistance, and has a night IR cut filter. The camera specifications ensure high-quality footage capture, even in varying environmental conditions and low-light intensities. The footage was recorded continuously for 24 h. Since counting ants’ activity at every minute was challenging, we selected only the first 10 min of each hour within the 24-h day to count entry-exit activity. All recorded data were stored in the camera’s internal hard disk, providing a secure and easily accessible storage method for subsequent analysis. The observed entry-exit data were compared and tallied for nest-wise, day-wise, and season-wise comparisons.

On a daily basis, a total of 24 data points were recorded for every day. Data were collected over a period of 30 days for each of the 9 nests (nests 1-9), and as a result, 720 data points (30 × 24) have been compiled for each nest for our investigation (6,480 data points for 9 nests).

Data Analysis

Determination of Nest Entry and Exit Fidelity

We left out the first 2 days of foraging activity data for nest 1 as activity levels were unusually high due to nest-building. In addition, there was a single occurrence of a missing value, which was observed for nest 2 on day 20, during hour 2, for the exit counts. Linear interpolation was used to estimate this missing value in the data set. Otherwise, all analyses and figures included in this study utilized all available data, unless otherwise specified.

The reliability of our counts of entries and exits was verified by first computing the ratio between entry and exit counts for each day in each nest. We calculated the median daily ratio for each nest, which served as an indicator of the level of disagreement between the recorded entry and exit counts.

Measuring Overall Foraging Activity

To assess the general behavior of O. smaragdina under different environmental conditions or for specific time periods throughout the day, we quantified the aggregate number of entries or exits within a 24-h time frame. This measurement is represented by the sum of the entry or exit counts within the 24-h period. We refer to this measurement as the area under the curve (AUC). The minimum value of AUC per 24-h period is 0, and a larger AUC value indicates greater foraging activity than a lower value.

Measuring Periodicity of Foraging Activity

To assess the periodicity of foraging activity under different environmental conditions, we employed the continuous wavelet transform (CWT) function available in the Wavelet Toolbox version 6.2 of MATLAB R2022b (The MathWorks, Inc., 2022). The values in each column of the 2-dimensional (2D) magnitude scalogram were normalized to a range of −1 to 1. The period in which the highest value in the column was observed was determined and designated as the dominant periodicity for that specific time point. In this study, we conducted a comparison between the median dominant periodicity observed on rainy days and non-rainy days (either dry season or rainy season). In addition, we examined the periodicity of foraging activity under conditions of constant predation. We used the Mann-Whitney U test, implemented by Virtanen et al. (2020) from the Python package SciPy version 1.7.3, to compare the dominant periodicities from non-rain and non-predator conditions to that of when they are present.

Assessing Deviation From Normal, Non-Rainy Foraging Activity and Rebound From Rain

Euclidean distance was employed to assess the structural similarity between 2 specific days, d₁ and d₂, where d₁ and d₂ contain hourly counts of entries into a nest. To assess the structural similarity between days characterized by the presence or absence of rain, we first computed the median count of nest entries per hour on no rain days. This allowed us to establish a comprehensive 24-h foraging activity template for each nest, specifically for days without rain. Subsequently, the Euclidean distance was computed between the nest-specific template and all days encompassing both rain and non-rain conditions for the given nest. All Euclidean distances between the nest-specific template and all days within the nest were normalized by the maximum Euclidean distance. A step-by-step example of this analysis is shown in Supplemental Figure 2. We used the Mann-Whitney U test to compare the Euclidean distances on days with rain with days without rain.

We assessed the rebound of foraging activity after the first rainy period by comparing the normalized Euclidean distance of the first non-rainy day after the first period of rain (denoted day 0) to the dry days immediately following day 0 using the Wilcoxon one-sample signed-rank test implemented by Virtanen et al. (2020) from the Python package SciPy version 1.7.3. While some nests have more than 1 rainy period, we chose to only evaluate the rebound after the first rainy period as the number of days following the second period of rain is often small.

Identifying Phase-Shifted Foraging Activity

To ascertain a transition from diurnal to nocturnal behavior, we partitioned the hours of the night (specifically, from 1800 to 0500 h) and the hours of the day (specifically, from 0600 to 1700 h), subsequently calculating the AUC for each time period on days without rainfall. A 1-sided Mann-Whitney U test was performed to determine whether the AUC during nighttime was significantly higher than the AUC during daytime for nests exposed to the dry season, rainy season, and predation.

Results

Entry-Exit Fidelity of Oecophylla

The time series data depicting the entries and exits for nests 1 through 9 are presented in Figure 1a-1i. The study revealed that the median disagreement between entries and exits, as measured by the median daily ratio of entries to exits, was 0.99 across all nests (Figure 1j-1r). Given the minimal discrepancy observed between entry and exit counts, we exclusively utilized entry counts for all subsequent analyses.

Figure 1.

Hourly entry counts, hourly exit counts, and daily entry-exit differences for nests 1 through 6. (a-i) Time series of hourly entry and exit counts for nests 1 through 9. Days with rain are highlighted in blue, and days with predation are highlighted in purple. (j-r) Differences between total daily entries and exits for each nest over all available days are shown; positive differences are highlighted in green and negative differences in red. The median entry/exit ratio across all days in a nest is denoted “Ratio”; the total sum of daily entry-exit differences across all days for a nest is denoted “Area.”

Rain Masking Reduces Oecophylla Foraging Activity

It was observed that the AUC of foraging activity on days characterized by rainfall was significantly lower compared to the AUC on days without rainfall for all nests subject to both rainy and non-rainy days (Table 1; Figure 2a-2f). The median AUC for days with rain was 710.5 across nests 1 through 6, while the median AUC for days without rain was 1578.5 (Median_{no rain} [MAD] = 1,595.0 [262.0], Median_rain [MAD] = 755.0 [210.0], n_{no rain} = 141, n_rain = 37, U = 4865.5, p < 0.00001; Figure 2g). We further compared the foraging activity on days without rain from the dry season (nests 1, 2, and 3) with days without rain from the rainy season (nests 4, 5, and 6; Figure 2h): we found a significant seasonal difference in AUC between the dry and rainy season, with the dry season having a greater AUC (Median_{dry season, no rain} [MAD] = 1,883.5 [429.5], Median_{rainy season, no rain} [MAD] = 1,491.0 [103.0], n_{dry season, no rain} = 82, n_{rainy season, no rain} = 59, U = 3,337.5, p = 0.00012; Figure 2h). The same seasonal separation of days with rain showed no difference in AUC (Median_{dry season, rain} [MAD] = 632.5 [262.0], Median_{rainy season, rain} [MAD] = 814.0 [169.0], n_{dry season, rain} = 6, n_{rainy season, rain} = 31, U = 70.0, p = 0.36; Figure 2i).

Table 1.

Comparison of foraging activity (AUC) between rainy and non-rainy days.

	Median_{no rain} (MAD)	Median_rain (MAD)	n _{no rain}	n _rain	U	p
Nest 1	N/A; no days with rain.
Nest 2	1,881.0 (50.0)	986.0 (78.0)	27	3	78.0	0.011
Nest 3	1,315.0 (49.0)	379.0 (17.0)	27	3	81.0	* 0.0057
Nest 4	1,494.5 (104.5)	742.0 (64.0)	18	12	203.5	*** 0.000058
Nest 5	1,423.0 (81.0)	937.0 (178.0)	21	9	187.0	*** 0.000031
Nest 6	1,587.0 (65.0)	633.5 (233.5)	20	10	161.0	* 0.0078

Abbreviation: MAD = median absolute deviation. Reported statistics are from the 2-sided Mann-Whitney U test, and reported p values are untransformed; significance level (*) determined after Bonferroni correction for 5 comparisons: (i.e., p < 0.05, 0.01, 0.001, and 0.0001 are assessed as *p < 0.01. **p < 0.002. ***p < 0.0002. ****p < 0.00002).

Figure 2.

AUC comparison of rainy days to non-rainy days by nest followed by a comparison of seasonal differences in AUC. (a-f) Time series of 24-h segments of entry counts into nests 1 through 6 on days without rain (gray) and days with rain (where applicable, light blue). Opaque black and blue lines indicate the median 24-h time series of entries for days without rain and days with rain, respectively. (g) Median 24-h time series of entry counts across individual days in nests 1 through 6 on days without rain (black) and days with rain (blue). The median absolute deviation of the median time series is shown in gray and light blue, respectively. The p value for the AUC of days with rain compared with days without rain is indicated on each subplot, where applicable. (h) Comparison of AUC for days without rain from the dry season (light orange) to the rainy season (light purple). The opaque orange line indicates the median 24-h time series of the individual days for nests subject to the dry season (nest group 1). The opaque purple line indicates the median 24-h time series of the individual days for nests subject to the rainy season (nest group 2). (i) Comparison of AUC for days with rain from the dry season to the rainy season with the same color scheme as (h). The p value for the comparison of AUCs from the dry season compared with AUCs from the rainy season is indicated on the subplots. *p < 0.01. **p < 0.002. ***p < 0.0002. ****p < 0.00002 using the 2-sided Mann-Whitney U test.

Rain Masking Destabilizes Periodicity of Oecophylla Foraging Activity

On days without rain, we found that the median dominant periodicity of O. smaragdina foraging activity in nests 1 through 6 occurred at 23.97 h (Table 2; Figure 3a-3f). On days with rain, the median dominant periodicity significantly deviated from the median 23.97 hours for nests 2 through 5 (Table 2).

Table 2.

Comparison of dominant periodicities between rainy and non-rainy days.

	Median_{no rain} (MAD)	Median_rain (MAD)	n _{no rain}	n _rain	U	p
Nest 1	N/A; no days with rain.
Nest 2	23.97 (0.0)	23.63 (0.34)	648	72	28,138.0	** 0.0013
Nest 3	23.97 (0.0)	24.32 (0.69)	648	72	17,065.0	**** <0.00002
Nest 4	23.97 (0.0)	10.37 (1.61)	432	288	108,006.5	**** <0.00002
Nest 5	23.97 (0.0)	21.98 (3.42)	504	216	78,882.5	**** <0.00002
Nest 6	23.97 (0.0)	25.03 (3.89)	480	240	29,916.5	**** <0.00002

Abbreviation: MAD = median absolute deviation. Reported statistics are from the 2-sided Mann-Whitney U test, and reported p-values are untransformed; significance level (*) determined after Bonferroni correction for 5 comparisons: (i.e., p < 0.05, 0.01, 0.001, and 0.0001 are assessed as *p < 0.01. **p < 0.002. ***p < 0.0002. ****p < 0.00002).

Figure 3.

Periodicity of foraging activity and similarity of rainy days to non-rainy days. (a-f) The CWT was applied to all available entry data for nests 1 through 6, respectively. The maximum normalized magnitude for each column in the magnitude scalogram is shown in yellow. Time periods with rain are boxed and highlighted in blue. (g) Euclidean distance between a 24-h, non-rainy foraging activity template for each nest and each day in that same nest. Euclidean distance before day 0 was compared to Euclidean distance on and after day 0 for each nest using the 2-sided Mann-Whitney U test, the p values of which are indicated in the subplot.

We measured the deviation of foraging activity during rain and rebound of foraging activity after rain by calculating the Euclidean distance between each day in a nest to the 24-h, non-rainy foraging activity template for each nest. We aligned nests 2 through 6 by the first day without rain after the first series of rainy days (referred to as day 0). The normalized Euclidean distance between the 24-h template and all days in each nest was significantly larger on days with rain compared with days without rain for most nests (Table 3; Figure 3g). We found that the Euclidean distance on day 0 was significantly different than the following days without rain (Table 4; Figure 3g).

Table 3.

Comparison of normalized Euclidean distance between each day and its 24-h, non-rainy foraging activity template for each nest.

	Median_{no rain} (MAD)	Median_rain (MAD)	n _{no rain}	n _rain	U	p
Nest 1	N/A; no days with rain.
Nest 2	0.12 (0.019)	0.81 (0.031)	27	3	10.0	0.033
Nest 3	0.30 (0.043)	0.92 (0.084)	27	3	0.0	** 0.00049
Nest 4	0.34 (0.13)	0.90 (0.058)	18	12	0.0	**** <0.00002
Nest 5	0.096 (0.016)	0.72 (0.086)	21	9	1.0	*** 0.000026
Nest 6	0.11 (0.033)	0.88 (0.10)	20	10	0.0	**** <0.00002

Abbreviation: MAD = median absolute deviation. Reported statistics are from the Mann-Whitney U test, and reported p values are untransformed; significance level (*) determined after Bonferroni correction for 5 comparisons: (i.e., p < 0.05, 0.01, 0.001, and 0.0001 are assessed as *p < 0.01. **p < 0.002. ***p < 0.0002. ****p < 0.00002).

Table 4.

Comparison of normalized Euclidean distance between the first non-rainy (rebound) day following a period of rain and the successive (post-rebound) days.

	Median_{no rain} (MAD)	Rebound Value	n _{no rain}	T	p
Nest 1	N/A; no days with rain.
Nest 2	0. 12 (0.018)	0. 65	25	9.0	**** <0.00002
Nest 3	0.30 (0.035)	0.34	24	37.0	** 0.00065
Nest 4	0.44 (0.052)	0.45	12	27.0	0.38
Nest 5	0.097 (0.014)	0.20	18	18.0	** 0.0019
Nest 6	0.12 (0.045)	0.20	16	11.0	** 0.0017

Abbreviation: MAD = median absolute deviation. Reported statistics are from the 1-sample Wilcoxon signed-rank test, and reported p values are untransformed; significance level (*) determined after Bonferroni correction for 5 comparisons: (i.e., p < 0.05, 0.01, 0.001, and 0.0001 are assessed as *p < 0.01. **p < 0.002. ***p < 0.0002. ****p < 0.00002).

Predation Masked Overall Foraging Activity and Shifts Foraging Activity Toward Nighttime

We observed that nests under predation (nest group 3) appeared to exhibit greater foraging activity at nighttime relative to daytime compared to nests without predators (Figure 4a). Like non-rainy days in nest groups 1 and 2, we found that nest group 3 maintains an approximately daily rhythm of foraging activity in the absence of predators with a median dominant periodicity of 23.97 h for all nests (Table 5; Figure 4b-4d). On days with predators, the median dominant periodicity significantly deviated from days without predators for only nest 7 (Table 5). We found that daily AUC in nest group 3 in the absence of predators was significantly smaller compared with AUC on days without rain in nest group 1 and nest group 2 (p < 0.0001 for both and Bonferroni-corrected; Figure 4e-4g). Nighttime foraging activity significantly increased relative to daytime foraging activity for nest group 3 when under predation (Table 6; Figure 4g). For nests not under predation, the opposite was true: we found greater foraging activity during daytime than nighttime for both nest groups 1 and 2 during non-rain conditions (Table 6; Figure 4e and 4f).

Figure 4.

Overall foraging activity and periodicity of foraging activity for nests under predation and comparison to other nests. (a) Entry counts of nests under predation (nest group 3: nests 7 through 9) and all other nests (1 through 6). (b-d) The CWT was applied to all available entry data for nests 7, 8, and 9, respectively. The maximum normalized magnitude for each column in the magnitude scalogram is shown in yellow. Time periods with predators present are boxed and highlighted in purple. (e) 24-h time series segments of entry counts for nest group 1 (nests 1 through 3) on days without rain during the dry season (gray). The opaque black line indicates the median 24-h time series of the individual days for nest group 1. Median AUC is provided. (f and g) Show data in the same manner as (e) but for nest group 2 (nests 4 through 6) during the rainy season and nest group 3 under predation, respectively. ****p < 0.0002 using the 1-sided Mann-Whitney U test to determine whether nighttime foraging activity exceeds daytime foraging activity.

Table 5.

Comparison of dominant periodicities in the absence and presence of predation.

	Median_{no predation} (MAD)	Median_predation (MAD)	n _{no predation}	n _predation	U	p
Nest 7	23.97 (0.0)	23.63 (0.69)	312	408	86,102.5	**** <0.000033
Nest 8	23.97 (0.34)	23.97 (1.06)	384	336	64,847.0	0.90
Nest 9	23.97 (0.0)	23.97 (0.35)	288	432	58,589.0	0.16

Abbreviation: MAD = median absolute deviation. Reported statistics are from the Mann-Whitney U test, and reported p values are untransformed; significance level (*) determined after Bonferroni correction for 3 comparisons: (i.e., p < 0.05, 0.01, 0.001, and 0.0001 are assessed as *p < 0.017. **p < 0.0033. ***p < 0.00033. ****p < 0.000033).

Table 6.

Comparison of foraging activity (AUC) between nighttime and daytime.

	Median_nights (MAD)	Median_days (MAD)	n _nights	n _days	U	p
Dry season	475.5 (187.5)	1,339.5 (122.5)	82	82	422.0	>0.99
Rainy season	359.0 (40.0)	1,114. (107.0)	59	59	0.0	>0.99
Predation	365.0 (64.0)	98.0 (40.0)	49	49	2,156.5	**** <0.0000333

Abbreviation: MAD = median absolute deviation. Reported statistics are from the Mann-Whitney U test, and reported p values are untransformed; significance level (*) determined after Bonferroni correction for 3 comparisons: (i.e., p < 0.05, 0.01, 0.001, and 0.0001 are assessed as *p < 0.017. **p < 0.0033. ***p < 0.00033. ****p < 0.000033).

Discussion

The overall foraging activity of O. smaragdina is significantly masked under conditions of rain, whereas it appears to phase shift in response to diurnal predation; furthermore, under rainy conditions, all nests exhibited a significant deviation from their typical, non-rain daily periodicity of foraging activity. The rain-induced disruption in overall foraging activity appeared consistent with circadian masking because non-rain levels of foraging activity immediately rebound the day following a period of rain. Larger and smaller nests may encounter distinct energetic challenges, allowing for useful comparisons of nest behavior following rain-induced foraging suppression. Longer rain periods may also reveal thresholds of energetic stores in future observations. Interestingly, most nests under predation maintain a roughly daily rhythm of foraging activity but exhibit a phase shift toward greater activity during nighttime relative to daytime hours; therefore, ants under predation may be exhibiting niche plasticity (Smarr et al., 2013) during which they adjust their preferred foraging hours to avoid a diurnal predator.

Insects must in some way “calculate” a cost-benefit analysis between nocturnal and diurnal lifestyles due to the difference in food availability, climatic condition, and the threat of predators during the time (Metcalfe et al., 1999). Foraging patterns and intensity are typically determined by the availability of food resources across their habitat (Westreich et al., 2023). This may be consistent with our observation that seasonal variations in daily foraging activity were observed on rain-free days, with less overall foraging activity during the rainy season. The deviation from non-rainy, daily foraging activity also appears greater during rainy days from the rainy season (nests 4-6) compared with rainy days from the dry season (nests 2 and 3). The greater disruption in periodicity during the rainy season may be driven by greater frequency or intensity of rainfall than in the dry season.

Insects’ foraging strategies generally differ according to the nature of the species’ fitness and the existing threats and competitions in their roaming area. For example, some bees such as Proxylocopa olivieri exhibit bimodal foraging patterns at dawn and dusk because it is advantageous to them to collect nectar (Gottlieb et al., 2005). Many types of ants, such as harvesters (Pogonomyrmex sp.) (MacKay and MacKay, 1989), leaf-cutters (Atta laevigata) (Viana et al., 2004), desert ants (Cataglyphis and Ocymyrmex sp.) (Wehner and Wehner, 2011), and even weaver ants (O. smaragdina), display bimodal foraging behavior in dawn and dusk due to favorable climatic conditions. In our study, we also noticed that some nests subject to the rainy season appear to exhibit bimodal foraging activity; we did not attempt to describe this in detail from a small number of nests, and instead leave it to future work.

The adaptive ability for selection on the collective behavior of ants, which governs foraging activity, may influence how the population adapts to climatic change caused by rain (Sundaram et al., 2022). The greater destabilization of the daily rhythm on rainy days during the rainy season (nest group 2) likely has to do with the greater duration (i.e., continuous rainfall) or intensity of rain. Ants tend to be less active on rainy days, which suggests that they have sufficient food reserves in their nest and do not need to search for more. This is supported by their swift return to typical foraging patterns following a period of rain rather than engaging in compensatory foraging activity.

The foraging activity pattern of African savanna termites exhibits seasonal variations that are influenced by the gradient of rainfall. They are more active during the rainy and transition seasons than in the dry ones (Davies et al., 2015). Based on their foraging activity pattern, 2 types of Mediterranean ant communities are distinguished based on their heat tolerance capabilities (Cros et al., 1997). The foraging activity rhythm of the Neotropical forest-dwelling ant, Odontomachus chelifer, varies with temperature and prey availability (Raimundo et al., 2009). Oecophylla’s overall foraging activity during non-rainy days from the dry season is significantly greater compared with the wet season.

Birds practice anting (ant-eating) for several reasons; some theories suggest that anting helps birds to control ectoparasites by utilizing the formic acid secreted by ants (Clayton et al., 2010; Ehrlich et al., 1986). Anting assists birds in cleaning and preserving their plumages (Morozov, 2015; Powlesland, 1982), aids in maturation, and relieves skin irritation (Potter, 1970; Revis, 2002). The jungle babbler (A. striata) and the Indian myna (A. tristis) forage during the day (Mahabal and Vaidya, 1989; Yambem and Jain, 2023), chirp frequently (Ali and Ripley, 1983; Gaston, 1977), and often forage in pairs or small groups (Federspiel et al., 2017). Since they frequently predate O. smaragdina during the day, it is possible to hypothesize that the ant makes an energetic trade-off to deal with the predation pressure of birds and shifts their foraging activity to nighttime. Insects like ants and agriculturally significant species adapt to environmental challenges, such as predation and climate shifts, by altering their behavior. This adaptation reflects an energetic trade-off, where they adjust activities like foraging or shift to nocturnal behavior to enhance survival. For instance, ants reduce foraging during rain to conserve energy, while becoming nocturnal helps evade daytime predators. However, these adaptations incur potential energy costs, and developing a framework to predict unsustainable changes due to energy limitations is crucial. Understanding these trade-offs aids in predicting extinction risks and guiding conservation efforts, supporting these insects’ vital roles in ecosystems and agriculture. Ants often possess the ability to sense predators like birds, prompting a decrease in foraging activity as a defense mechanism. This response could be triggered by cues indicating the presence of danger, leading to fewer foragers due to predation. This decrease in foragers may impact the nest’s food-gathering capacity and overall ant population. It highlights how predation can influence ant behavior and colony dynamics trade-offs.

Understanding the relationships between predators and prey as well as the effects of weather are essential for maintaining biodiversity since climate change alters the historical patterns to which species have adapted. Our study examined both factors in a single species and found that the species had a repertoire of responses that are likely governed by energetic trade-offs.

Supplemental Material

sj-tif-1-jbr-10.1177_07487304241233778 – Supplemental material for Divergent Circadian Foraging Strategies in Response to Diurnal Predation Versus Persistent Rain in Asian Weaver Ant, Oecophylla smaragdina, Suggest Possible Energetic Trade-offs

Supplemental material, sj-tif-1-jbr-10.1177_07487304241233778 for Divergent Circadian Foraging Strategies in Response to Diurnal Predation Versus Persistent Rain in Asian Weaver Ant, Oecophylla smaragdina, Suggest Possible Energetic Trade-offs by Avishek Dolai, Severine Soltani, Benjamin Smarr and Amlan Das in Journal of Biological Rhythms

Supplemental Material

sj-tif-2-jbr-10.1177_07487304241233778 – Supplemental material for Divergent Circadian Foraging Strategies in Response to Diurnal Predation Versus Persistent Rain in Asian Weaver Ant, Oecophylla smaragdina, Suggest Possible Energetic Trade-offs

Supplemental material, sj-tif-2-jbr-10.1177_07487304241233778 for Divergent Circadian Foraging Strategies in Response to Diurnal Predation Versus Persistent Rain in Asian Weaver Ant, Oecophylla smaragdina, Suggest Possible Energetic Trade-offs by Avishek Dolai, Severine Soltani, Benjamin Smarr and Amlan Das in Journal of Biological Rhythms

Footnotes

Acknowledgements

The authors acknowledge the HOD, Department of Zoology, University of Calcutta, for providing the space for conducting the experiments, including UGC DRS-SAP and DST-FIST and thanks to Center for Circadian Biology at University of California San Diego for connecting the authors. A.D. thanks the UGC, Government of India for providing research fellowship (Sanction No 751/(CSIR-UGC NET JUNE 2018); dated: April 15, 2019). S.S. was supported by a grant from the National Institutes of Health, USA (NIH grant T32GM139790).

Conflict Of Interest Statement

The authors have no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

ORCID IDs

Avishek Dolai

Benjamin Smarr

Amlan Das

Notes

References

Ali

Ripley

(1983) Handbook of the birds of India and Pakistan. Compact edition. Mumbai (India): Oxford University Press and BNHS.

Ambika

Nalini

(2019) Nest composition, inward and outward flows of Oecophylla smaragdina Fabricius (Hymenoptera: Formicidae) in selected fruit crops. Plant Arch 19:2781-2784.

Brown

Braithwaite

(2005) Effects of predation pressure on the cognitive ability of the poeciliid Brachyraphis episcopi. Behav Ecol 16:482-487.

Button

Seifert

Chow

Araujo

(2020) Dynamics of skill acquisition: an ecological dynamics approach. 2nd ed. Champaign (IL): Human Kinetics Publishers.

Cane

Schiffhauer

Kervin

(1996) Pollination, foraging, and nesting ecology of the leaf-cutting bee Megachile (Delomegachile) addenda (Hymenoptera: Megachilidae) on cranberry beds. Ann Entomol Soc Am 89:361-367.

Chua

K-O

Liew

YJM

See-Too

W-S

Tan

J-Y

Yong

H-S

Yin

W-F

Chan

K-G

(2022) Formicincola oecophyllae gen. nov. sp. nov., a novel member of the family Acetobacteraceae isolated from the weaver ant Oecophylla smaragdina. Anton Leeuw 115:995-1007.

Clayton

Koop

Harbison

Moyer

Bush

(2010) How birds combat ectoparasites. Open Ornithol J 3:41-71.

Crall

Gravish

Mountcastle

Kocher

Oppenheimer

Pierce

Combes

(2018) Spatial fidelity of workers predicts collective response to disturbance in a social insect. Nat Commun 9:1201.

Cros

Cerdá

Retana

(1997) Spatial and temporal variations in the activity patterns of Mediterranean ant communities. Ecoscience 4:269-278.

10.

Crozier

Newey

Schluens

Robson

SKA

(2010) A masterpiece of evolution–Oecophylla weaver ants (Hymenoptera: Formicidae). Myrmecol News 13:57-71.

11.

Das

Gordon

(2023) Biological rhythms and task allocation in ant colonies. Curr Opin Insect Sci 58:101062.

12.

Davies

Eggleton

van Rensburg

Parr

(2015) Seasonal activity patterns of African savanna termites vary across a rainfall gradient. Insectes Soc 62:157-165.

13.

Ehrlich

Dobkin

Wheye

(1986) The adaptive significance of anting. Auk 103:835.

14.

El Keroumi

Naamani

Outoukarte

Abdelhi

Dahbic

(2022) Biocontrol and Circadian Rhythms of Medfly larvae emergence and Monomorium subopacum ant foraging along the fruiting season in the Argan forest (Morocco). Int J Pest Manag. Published online November 7. doi:10.1080/09670874.2022.2141909

15.

Evans

Smith

Raine

(2021) Odour learning bees have longer foraging careers than non-learners in a natural environment. Front Ecol Evol 9:676289.

16.

Farji-Brener

Dalton

Balza

Courtis

Lemus-Domínguez

Fernández-Hilario

Cáceres-Levi

(2018) Working in the rain? why leaf-cutting ants stop foraging when it’s raining. Insectes Soc 65:233-239.

17.

Federspiel

Garland

Guez

Bugnyar

Healy

Güntürkün

Griffin

(2017) Adjusting foraging strategies: a comparison of rural and urban common mynas (Acridotheres tristis). Anim Cogn 20:65-74.

18.

Friedman

(2019) Behavioral, physiological, and transcriptomic variation among colonies of the Red harvester ant (Pogonomyrmex barbatus). Stanford (CA): Stanford University.

19.

Fujioka

Abe

Okada

(2019) Ant activity-rest rhythms vary with age and interaction frequencies of workers. Behav Ecol Sociobiol 73:1-12.

20.

Gaston

(1977) Social behaviour within groups of jungle babblers (Turdoides striatus). Anim Behav 25:828-848.

21.

Gottlieb

Keasar

Shmida

Motro

(2005) Possible foraging benefits of bimodal daily activity in Proxylocopa olivieri (Lepeletier)(Hymenoptera: Anthophoridae). Environ Entomol 34:417-424.

22.

Hanisch

Blanco

Tubaro

Suarez

(2023) Spatial fidelity and uniform exploration in the foraging behaviour of a giant predatory ant. Anim Behav 203:63-73.

23.

Hodgson

(1955) An ecological study of the behavior of the leaf-cutting ant Atta cephalotes. Ecology 36:293-304.

24.

Hölldobler

Wilson

(1990) The ants. Cambridge (MA): Harvard University Press.

25.

Jürgen Stelzer

Stanewsky

Chittka

(2010) Circadian foraging rhythms of bumblebees monitored by radio-frequency identification. J Biol Rhythms 25:257-267.

26.

Kronfeld-Schor

Bloch

Schwartz

(2013) Animal clocks: when science meets nature. Proc Royal Soc B Biol Sci 280:20131354.

27.

Lehue

Detrain

(2020) Foraging through multiple nest holes: an impediment to collective decision-making in ants. PLoS One 15:e0234526.

28.

Lone

(2023) Rest of the carpenter ant Camponotus compressus undergoes a drastic change after mating. Insectes Soc 70:97-103.

29.

MacKay

(1989) Diurnal foraging patterns of Pogonomyrmex harvester ants (Hymenoptera: Formicidae). Southwest Nat 34:213-218.

30.

Mahabal

Vaidya

(1989) Diurnal rhythms and seasonal changes in the roosting behaviour of Indian Myna Acridotheres tristis (Linnaeus). Proc Anim Sci 98:199-209.

31.

The MathWorks, Inc. (2022) Wavelet toolbox version 6.2 (R022b) [Computer software]. The Mathworks, Inc. https://www.mathworks.com.

32.

Mersch

Crespi

Keller

(2013) Tracking individuals shows spatial fidelity is a key regulator of ant social organization. Science 340:1090-1093.

33.

Metcalfe

Fraser

Burns

(1999) Food availability and the nocturnal vs. diurnal foraging trade-off in juvenile salmon. J Anim Ecol 68:371-381.

34.

Mildner

Roces

(2017) Plasticity of daily behavioral rhythms in foragers and nurses of the ant Camponotus rufipes: influence of social context and feeding times. PLoS One 12:e0169244.

35.

Miyazaki

Nisimura

Numata

(2014) Circannual rhythms in insects.Circannual rhythms in insects. In: Numata

Helm

, editors. Annual, lunar, and tidal clocks: patterns and mechanisms of nature’s enigmatic rhythms. London (England): Springer. p. 333-350.

36.

Morozov

(2015) Why do birds practice anting? Biol Bull Rev 5:353-365.

37.

Mukhopadhyay

Sannigrahi

(1993) Rearing success of a polyphagous predator Geocoris ochropterus (Hem.: Lygaeidae) on preserved ant pupae of Oecophylla smaragdina. Entomophaga 38:219-224.

38.

O’Donnell

Lattke

Powell

Kaspari

(2021) Diurnal and nocturnal foraging specialisation in Neotropical army ants. Ecol Entomol 46:352-359.

39.

Orr

(1992) Parasitic flies (Diptera: Phoridae) influence foraging rhythms and caste division of labor in the leaf-cutter ant, Atta cephalotes (Hymenoptera: Formicidae). Behav Ecol Sociobiol 30:395-402.

40.

Page

(1989) Masking in invertebrates. Chronobiol Int 6:3-11.

41.

Philpott

Greenberg

Bichier

(2005) The influence of ants on the foraging behavior of birds in an agroforest 1. Biotropica 37:468-471.

42.

Pinter-Wollman

Hubler

Holley

J-A

Franks

Dornhaus

(2012) How is activity distributed among and within tasks in Temnothorax ants? Behav Ecol Sociobiol 66:1407-1420.

43.

Potter

(1970) Anting in wild birds, its frequency and probable purpose. Auk 87:692-713.

44.

Powlesland

(1982) Body-maintenance activities of the South Island Robin. Emu 82:296-304.

45.

Raimundo

Freitas

Oliveira

(2009) Seasonal patterns in activity rhythm and foraging ecology in the neotropical forest-dwelling ant, Odontomachus chelifer (Formicidae: Ponerinae). Ann Entomol Soc Am 102:1151-1157.

46.

Reeves

Moreau

(2019) The evolution of foraging behavior in ants (Hymenoptera: Formicidae). Arthropod Syst Phylogeny 77:351-363.

47.

Revis

(2002) Anting behavior in birds: ant selection and effect of ant chemistry on feather ectoparasites [Ph.D. thesis]. [Norfolk (VA)]: Old Dominion University.

48.

Rietveld

Minors

Waterhouse

(1993) Circadian rhythms and masking: an overview. Chronobiol Int 10:306-312.

49.

Roces

Nunez

(1996) A circadian rhythm of thermal preference in the ant Camponotus mus: masking and entrainment by temperature cycles. Physiol Entomol 21:138-142.

50.

Rockwood

(1975) The effects of seasonality on foraging in two species of leaf-cutting ants (Atta) in Guanacaste Province, Costa Rica. Biotropica 7:176-193.

51.

Sharma

Lone

Goel

(2004) Clocks for sex: loss of circadian rhythms in ants after mating? Naturwissenschaften 91:334-337.

52.

Sharma

Lone

Goel

Chandrashekaran

(2004) Circadian consequences of social organization in the ant species Camponotus compressus. Naturwissenschaften 91:386-390.

53.

Sharma

Lone

Mathew

Goel

Chandrashekaran

(2004) Possible evidence for shift work schedules in the media workers of the ant species Camponotus compressus. Chronobiol Int 21:297-308.

54.

Smarr

Schwartz

Wotus

de la Iglesia

(2013) Re-examining “temporal niche.” Integr Comp Biol 53:165-174.

55.

Sundaram

Steiner

Gordon

(2022) Rainfall, neighbors, and foraging: the dynamics of a population of red harvester ant colonies 1988–2019. Ecol Monogr 92:e1503.

56.

Traniello

(1989) Foraging strategies of ants. Annu Rev Entomol 34:191-210.

57.

Vayssières

J-F

Sinzogan

Korie

Adandonon

(2011) Field observational studies on circadian activity pattern of Oecophylla longinoda (Latreille)(Hymenoptera: Formicidae) in relation to abiotic factors and mango cultivars. Int J Biol Chem Sci 5. [accessed 2024 Feb 19]. https://www.ajol.info/index.php/ijbcs/article/view/72153.

58.

Viana

Santos

Arruda

Santos

Fernandes

(2004) Foraging patterns of the leaf-cutter ant Atta laevigata (Smith)(Myrmicinae: Attini) in an area of cerrado vegetation. Neotrop Entomol 33:391-393.

59.

Vidhu

Evans

(2015) Ethnoentomological values of Oecophylla smaragdina (Fabricius). Curr Sci 109:572-579.

60.

Villalta

Oms

Angulo

Molinas-González

Devers

Cerdá

Boulay

(2020) Does social thermal regulation constrain individual thermal tolerance in an ant species? J Anim Ecol 89:2063-2076.

61.

Virtanen

Gommers

Oliphant

Haberland

Reddy

Cournapeau

Burovski

Peterson

Weckesser

Bright

, et al (2020) SciPy 1.0: fundamental algorithms for scientific computing in Python. Nat Methods 17:261-272.

62.

Wehner

(2011) Parallel evolution of thermophilia: daily and seasonal foraging patterns of heat-adapted desert ants: Cataglyphis and Ocymyrmex species. Physiol Entomol 36:271-281.

63.

Westreich

Tobin

(2023) Native solitary bee reproductive success depends on early season precipitation and host plant richness. Oecologia 201:965-978.

64.

Yambem

Jain

(2023) Temporal variation in the behaviour of a cooperatively breeding bird, Jungle Babbler (Argya striata). Trop Ecol 64:133-145.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

15.00 MB

3.23 MB