Burrow ambient temperature influences Helice crab activity and availability for migratory Red‐crowned cranes Grus japonensis

Abstract For migratory birds that specialize on particular benthic macroinvertebrate species, the timing of migration is critical since prey availability may be temporally limited and a function of local ambient temperature. Hence, variation in local ambient temperature can influence the diet composition of migrant birds, and, consequently, they may be constrained by which stopover and wintering sites they are able to utilize during periods of colder temperatures. Here, we use fecal analysis, observer‐based population counts, digital video recordings, and temperature data to test five predictions regarding the influence of local ambient temperature on the activity and availability of mudflat crabs—a key prey resource at three staging/wintering sites in eastern China, for migratory Red‐crowned cranes (Grus japonensis) and how this subsequently influences crane diet and use of wetland sites. Pearson's correlations and generalized linear models revealed that mudflat crabs became significantly more surface active with increasing burrow ambient temperature. Piecewise regression analysis revealed that crab surface activity was largely limited to a burrow ambient temperature threshold between 12 and 13℃ after which activity significantly increased. Crab activity declining temporally during the crane's autumn migration period but increased during spring migration. Crabs accounted for a significant proportion of crane diet at two of three sites; however, the frequency of crab remains was significantly different between sites, and between autumn and spring migration. Analyses of crane count data revealed a degree of congruence between the migration timing of Red‐crowned cranes with periods of warmer ambient temperature, and a significant, positive correlation between the percentage of crab remains in crane feces and site ambient temperature. Collectively, our data suggest that temperature‐related mudflat crab activity may provide an important time window for migratory Red‐crowned cranes to utilize critical stopover sites and the crabs’ food resources.

The Red-crowned crane (Grus japonensis) is a globally threatened migratory species (BirdLife International, 2016) that has undergone a severe population decline in China since 2000 (Su & Zou, 2012).
Extensive loss and degradation of its primary breeding grounds in northeast China (Su & Zou, 2012) and extensive tidal land reclamation, wetland habitat invasion by smooth cordgrass (Spartina alterniflora), and expansion of oilfield production at wintering and staging areas (Cao, Xu, Le, Zhu, & Cao, 2015;Ma et al., 2009;Wang et al., 2019) have all contributed to its population decline.
Red-crowned cranes were formerly considered to be opportunistic feeders (Lee, Jablonski, & Higuchi, 2007;Ma, Wang, & Tang, 1999); however, at least one of our previous works (Li et al., 2014) revealed that mudflat crabs (Helice tientsinensis) in Suaeda salsa mudflat habitats form a crucial part of the species' winter diet in the Yellow River Delta. Li et al. (2014) also found that cranes increase their intake of mudflat crabs during late February to mid-March, prior to their northward spring migration. This period coincides with increasing regional ambient temperature and the emergence of mudflat crabs from their winter hibernation burrows (Figure 1). Regional ambient temperature has been widely suspected to be a significant predictor of the timing of Red-crowned crane autumn and spring migration. In F I G U R E 1 Showing Red-crowned cranes foraging in Suaeda salsa saltmarsh (a), with mudflat crab presence at the burrow entrance (b) and remains of mudflat crabs found after recent foraging activity by Red-crowned cranes (c) addition, there are other large areas of similar S. salsa intertidal mudflats in two other crane stopover/wintering sites situated approximately 900km apart: the Liaohe River Delta (LRD) and Yancheng Nature Reserve (YNR) and mudflat crab populations occur at both these sites.
In this study, we use fecal analysis, observer-based population counts, digital video recordings, and temperature data to determine whether increased ambient temperature affords better crab availability for Red-crowned cranes during autumn and spring at three protected area stopover/wintering sites in eastern China. We predicted that: (1) mudflat crabs become more surface active with increasing burrow ambient temperatures; (2) crab surface activity would be limited to a temperature threshold; (3) crab activity would decline temporally during the crane's autumn migration period but increase during the spring migration period with increasing temperature; (4) crane diet will vary across different stopover/wintering sites and between autumn and spring; (5) cranes select and use staging/ wintering sites with mudflat crabs when ambient temperature provides them with the opportunity to do so.  Table S1. The weather conditions were typical for these sites during our survey periods. All three sites represent coastal wetland mosaics consisting of S. salsa saltmarshes, intertidal mudflats, reed marsh habitat, and artificial aquaculture fishponds Ma et al., 1999;Xia et al., 2017). In this region, the S. salsa saltmarsha preferred foraging habitat for migratory Red-crowned cranes (Li et al., 2014)-has been greatly eroded in recent years due to a combination of land reclamation, changes to natural hydrological regimes, and invasion by non-native smooth cordgrass. This latter threat is particularly prevalent in the southern YNR and also within the YRD, but to date, has not yet been recorded in the LRD (Wang et al., 2019).

| Crab activity and ambient temperature data
The activities of mudflat crabs around their burrows were filmed using two digital cameras (Xiaoyi 4K, Shanghai, China). We randomly selected up to two crab burrows per day to film, with the camera positioned 0.5 m above the ground, to ensure an unobstructed view to the burrow entrance. At YRD, we conducted additional monitoring of crab activity at burrow entrances. Every 2-3 days, we randomly selected a sampling area measuring 5 m × 10 m within the S. salsa habitat to conduct the experiment. Ten plots measuring 1 m × 1 m were randomly selected within the sample area. From each plot, we randomly selected 10 crab burrows and the location of which were marked with wooden poles. No sampling plot was repeatedly sampled, and the distance between plots was no less than 10 m. Before 9a.m. on each sampling day, we plugged the entrance of the burrows with mud from the immediate environment around the burrow entrance. We then checked each burrow 24 hr later to determine whether the burrow was open or closed and used these data as an indication of crab activity.

| Fecal sample collection and prey identification
We collected a total of 902 fresh fecal samples of Red-crowned cranes from their foraging or roosting sites between 2011 and 2015 (Table 1). Crane feces were easily distinguishable from that of other species by their large amorphous volume and always included large amounts of crab remains. On just a few occasions, we found Redcrowned cranes feeding in the same habitat with Common crane (Grus grus) or Siberian crane (Leucogeranus leucogeranus), and only on these occasions we did not collect any fecal samples to avoid any error with allocating samples to the different crane species. We limited the number of fecal samples to ≤3 samples collected from each foraging or roost site to reduce the potential for pseudo-replication.
Only newly defecated samples were collected from the ground using a sterilized spoon, and these were subsequently stored in a sample tube and taken back to the field station (<8 hr travel time) and stored in refrigerator at −20°C before analysis. remains in the total sample per season in each site were calculated to represent site and seasonal variations in crane diet composition.

| Counts of migratory Red-crowned crane
Red

| Statistical analyses
Crab activity datasets from all sites were pooled for the analyses without considering inter-site differences. To test our first prediction, that mudflat crabs become more surface active with increasing burrow ambient temperature, we first used Pearson's correlations to examine the relationships between maximum daily and mean daily burrow ambient temperature with the onset (time) of crab activity time, and with the mean percentage of active crab burrows. In addition, we fitted two generalized linear models (GLMs) with the percentage time of crab activity at burrow entrances and percentage time crab activity outside of burrows on the mudflat as the response variables, with site (YRD, LRD), season (spring, autumn), daily maximum burrow ambient temperature and mean burrow ambient temperature included as predictor variables. Since both burrow ambient temperature response variables were positively correlated (r = .899, p < .001), we built these variables into the different models separately. We ran the GLMs with Poisson error structure and logit link function using the glm function included in the MASS package. We examined Wald test z scores to make inferences about each parameter estimate.
To test our second prediction (that crab surface activity would be limited to an ambient temperature threshold), we used a piecewise linear regression model to explore the relationships between percentage time of crab surface activity (combining the activity time both on the burrow entrance and outside the burrow on the mudflat) and maximum or mean burrow ambient temperature using the R package segmented (Muggeo, 2008 Figure S1). Thus, mudflat crabs started to become active from their burrows earlier on days with warmer temperature. When all the burrow monitoring data were pooled, we found that the percentage of active crab burrows increased with increases in daily maximum burrow ambient temperature (r = .244, p = .193) and daily mean burrow ambient temperature (r = .296, p = .112) but these increases were not significantly correlated ( Figure S2).

| RE SULTS
All GLMs, whether constructed using the daily maximum or daily mean burrow ambient temperatures, revealed significant positive influence of daily burrow ambient temperature on the mudflat crab activity either at burrow entrances or outside of burrows (Table 2). This suggests that mudflat crab activity was largely determined by burrow ambient temperature. In addition, the GLM models showed that there was significant variation in crab activity between seasons and sites, with crab activity being significantly higher in LRD than that in YRD. However, there was no significant difference in the percentage of time spent outside burrows between sites when daily maximum temperature was included in the model (Table 2). Note: Statistically significant estimates and standard errors (SEs) are highlighted in bold. The reference categories for "site" and "season" are "YRD" and "spring," respectively. The temperature was recorded using a Tinytag Plus 2 temperature data logger (TGP-4520; Gemini Data Loggers, UK).

TA B L E 2
Parameter estimates (logodds) from GLMMs of daily temperature (maximum and mean) on crab activity F I G U R E 3 Relationships of crab activity and maximum daily or mean daily burrow ambient temperatures, with the thresholds identified by break points (a: 11.9 ± 2.2; b: 2.8 ± 2.4) estimated by piecewise linear regression models. The temperature was recorded using a Tinytag Plus 2 temperature data logger (TGP-4520; Gemini Data Loggers, UK) 1. Is crab surface activity limited to a temperature threshold?
The average burrow ambient temperature for the onset of crab activity was 12.8°C ± 4.3 (25%~75% range, 10.37 ~ 15.38; n = 52), which was significantly lower in the autumn (11.1°C ± 4.2) than in spring (14.2°C ± 3.8; t = 2.799, df = 50, p = .007). The break point estimation of the piecewise linear regression shown that the percentage of time that crabs were recorded as being active at the burrow entrance and also outside of the burrows increased when daily mean ambient temperature was 2.8 ± 2.4°C and the daily maximum ambient temperature reached 11.9 ± 2.2°C (Figure 3; Table S2).
1. Does crab activity decline temporally during the crane's autumn migration period but increase during the spring migration? We

| D ISCUSS I ON
In this study, we have shown how variation in burrow ambient temperature influences diet composition of the migratory Red-crowned crane population across three wintering/stopover sites and different migration seasons through its influence on crab prey availability. Mudflat crab activity was largely influenced by burrow ambient temperature (prediction 1) and crab activity significantly increased at a burrow ambient temperature threshold of 12 ~ 13°C (prediction 2). Crab activity tended to advance with warmer ambient temperature, with the percentage of active burrows declining in the autumn and increasing again during the spring migration (prediction 3).
Temperature-related crab availability also has a significant influence on variation in the amount of crab prey found in crane feces between different staging/wintering sites and migration seasons (prediction 4). Finally, we also found a significant, positive relationship between the percentage of crab remains in the crane feces and local daily ambient temperature, and a degree of congruence between local ambient temperature and the arrival and departure timing of Red-crowned cranes at our three sites. This suggests that temperature-depended mudflat crab activity may provide an important time window for migratory Red-crowned cranes (prediction 5).
Red-crowned cranes typically spend 4-5 months staging for winter in the Yellow Sea coastal region (Higuchi et al., 1998;Ma et al., 1999;Su & Zou, 2012). To maintain their daily energy requirements, cranes should rely on predictable high-energy prey resources at these sites. Similar patterns of temporal synchrony in migration timing and prey availability have been documented for several migratory shorebird species. For example, the seasonal spawning of horseshoe crab (Limulus polyphemus) at Delaware Bay, F I G U R E 5 Temporal and site variations in the percentage of prey remains of Red-crowned cranes during their annual migration and wintering in three important sites (north-south: LRD: Liaohe River Delta Nature Reserve; YRD: Yellow River Delta Nature Reserve; YNR: Yancheng Nature Reserve) along the Yellow Sea F I G U R E 6 Relationships between crab remains in the feces of Red-crowned cranes with daily maximum ambient temperature (a: °C) and daily mean ambient temperature (b: °C). The data of temperature were downloaded from website (http://www.tianq ihoub ao.com) USA, provides a critical nutrient-rich dietary resource (eggs) for migratory populations of red knot Calidris canutus, ruddy turnstone Arenaria interpres, sanderling C. alba, and semipalmated sandpiper C.
Although mudflat crabs were the dominant prey of Redcrowned cranes staging in the LRD during the autumn, there were almost no crab remains found in crane feces at this site during the spring migration. The absence of mudflat crabs in the crane's spring diet at LRD may be largely due to the lower burrow ambient temperature at LRD on their arrival that restricted crab activity/ availability ( Figure 7b). This also highlights the trade-off made by migratory cranes-trying to synchronize the timing of their migration to exploit a suitable prey resource, with the need to reach their breeding ground. Ideally, Red-crowned cranes should time their northward spring migration to arrive at LRD some 3-4 weeks later than currently documented, when ambient temperatures are sufficiently high to influence greater crab activity. Our data do reveal that cranes staging at LRD may be able to adapt to a degree to feed on other food resources during the spring, and elsewhere, there are some reports that wintering Red-crowned cranes utilize anthropogenic habitats such as rice field (Li et al., 2013;Ma et al., 1999;Wang, Li, Beauchampe, & Jiang, 2011). Studies have revealed that numerous wading bird species vary their prey choice in response to variation in the availability of their different prey species (Zwarts & Wanink, 1993). Whether crane use of anthropogenic habitats is linked to burrow ambient temperature and crab availability remains unknown. Further research is needed to determine the degree with which these and other food types, in particular, corn, which is used as a supplementary feeding strategy by the protected area F I G U R E 7 Relationship between the use of three staging/wintering sites by Red-crowned cranes and the site-specific variation in ambient temperature (LRD: Liaohe River delta; YRD: Yellow River Delta; YNR: Yancheng Nature Reserve). a denotes ambient temperature recorded using a Tinytag Plus 2 temperature data logger (TGP-4520; Gemini Data Loggers, UK) in the YRD; b denotes site variation in the daily maximum ambient temperature (Download: http://www.tianq ihoub ao.com). The dotted line represented the threshold value of crab activity temperature identified using piecewise linear regression models. The number of individual cranes in b represents the migratory pattern of cranes in Liaohe River Delta (LRD). The red and black colored arrows represent the date of the daily maximum temperature exceeding the threshold value of 11.9°C in the YRD (February10, 2015) and LRD (March 15, 2015) management at YNR and LRD, can nutritionally compensate for the lack of crab prey in their diet during pre-migration (e.g., daily energy requirements, fat accumulation).
The prevalence of mudflat crabs in the crane's diet might be related to the ease of catching slower moving individuals during days with ambient temperatures at around the threshold of 12 ~ 13°C, or by catching inactive individuals found warming themselves at burrow entrances. Research from NW Europe has shown that invertebrate prey availability for visually feeding wading birds is significantly reduced when sediment surface temperatures fall below 3-6°C and that fewer prey are available for migrant species at sites with much colder ambient temperature (Zwarts & Wanink, 1993). We lacked detailed data to determine whether cranes resort to feeding largely on crabs within their burrows or at burrow entrances either during colder days, or on days when ambient temperature reach above the threshold (as opposed to foraging more on surface active crabs).
Little is known about anti-predator behavior of mudflat crabs during crane migration seasons at our study sites, and further research is needed to identify the period when mudflat crabs are most vulnerable to predation by cranes and whether any such periods correlate with spatial patterns of habitat use (burrow vs. mudflat surface). assistance with data collection in the field and four anonymous reviewers' comments on the structure and language of the manuscript.

CO N FLI C T O F I NTE R E S T
None declared.