High fitness areas drive the aggregation of the sea urchin Mesocentrotus nudus

Sea urchins

Five hundred and forty M. nudus (45.27 ± 2.48 mm test diameter and spines) were transported from a local aquafarm to the Key Laboratory of Mariculture & Stock Enhancement in the North China’ s Sea, Ministry of Agriculture and Rural Affairs, Dalian Ocean University. Twelve of the 540 sea urchins were randomly selected and dissected before the experiments. No sexual maturation or spawning was found in the samples (N = 12). Then, the rest 528 sea urchins were temporarily cultured with aerated seawater at ambient temperature (20 ± 1.0 °C) for 15 days in a tank (length × width × height: 85  × 50  × 65 cm, ∼276 L) under the natural photoperiod (12L: 12D) for the aggregation experiments. They were fed S. japonica ad libitum every three days. Sea urchins was subsequently held without food for three days before the experiment started on July 26, 2019. The filtered natural seawater was changed every three days. The 528 sea urchins were used in two experiments, including 352 urchins for Experiment 1 (224 in the food test, 128 in the predator test) and another 176 urchins for Experiment 2 (112 in the food test, 64 in the predator test). No sea urchin was used more than once in the experiments.

Experiment 1

An acrylic rectangular tank divide into two parts (length × width × height: 42  × 31.5 × 17.5 cm of each part) was used to investigate how M. nudus aggregate in response to food. Fixed polyvinyl nets at 3 cm from both sides of the tank separated the tank into a food part and a sea urchin part (Figs. 1A, S1). We took the occupied area of each sea urchin (test diameter + spines = ∼4.5 cm) as a square (4.5 cm × 4.5 cm). Therefore, 56 squares (4.5 cm × 4.5 cm) were marked in the experimental set-up for sea urchins (8  × 7 squares) as the total area for sea urchins. Pieces of S. japonica (15 g) were placed in both sides of the tank for the food group, making the 14 squares nearest to the kelp the HFAs. Fourteen sea urchins were subsequently placed on the squares of the tank. No food and special HFAs was involved in the control group with 14 sea urchins (Figs. 1B, S1). Twenty-eight sea urchins were used in each replicate of the food text (14 for the food group, 14 for the control group). The food test was repeated eight times in darkness with different sea urchins and kelp (N = 8).

To further investigate whether aggregation of sea urchins depends on various resources or is only induced by food, we performed a test of negative stimulus using the predator C. japonica. Similar to the food tank, the experimental set-up was an equally divided identical tank (length × width × height: 42  × 31.5 × 17.5 cm of each part) with additional fixed nets at 18 cm from the short side to separate predators from sea urchins (Figs. 1C, S2). Thirty-two squares were marked as the place to put in sea urchins. Two C. japonica were placed in the other side of the nets after one-week without food, making the eight squares farthest from the crabs as the HFAs for the eight M. nudus of the predator group. The fitness of the eight M. nudus in the control group without predator was the same in all squares (Figs. 1D, S2). Sixteen sea urchins were used in each replicate of the predator test (8 for the predator group, 8 for the control group). The predator test was repeated eight times in darkness using different sea urchins and kelp/crabs (N = 8).

Sea urchins were placed in diagonal squares (without contacting each other) at the beginning of the experiment and then allowed to move (Fig. 1). The behaviors were recorded for 30 min using an infrared digital camera (Legria HF20; Canon, Tokyo, Japan). We changed the seawater and washed the experimental tanks for each trial.

Experiment 2

The HFAs were relatively scarce for M. nudus in Experiment 1 (the squares of HFAs: the area occupied by experimental sea urchins = 1:1), because sea urchins would inevitably aggregate after they all migrated to HFAs at the ratio of 1:1. Alternatively, the ratio of 2:1 was set as a sufficient HFAs condition (the squares of HFAs: the area occupied by experimental sea urchins = 2:1). Because sea urchins have sufficient space of HFAs, they would remain separated from each other even when they all moved to the HFAs. Experiment 2 was thus designed to observe the aggregation of sea urchins under the sufficient HFAs.

The same experimental set-up in Experiment 1 was used in Experiment 2. With the addition of 15 g kelp for both sides of the experimental set-up of the food group, the 14 squares remain the HFAs for the M. nudus (Fig. 1E). But the number of sea urchins was halved to seven to set the ratio 2:1 (14 squares of HFAs for seven sea urchins). There was no special HFAs for the seven M. nudus in the control group, because no food was involved (Fig. 1F). Fourteen sea urchins were used in each replicate (seven for the experimental group, 7 for the control group). The food test was repeated eight times with different sea urchins and kelp (N = 8).

Likewise, two crabs were placed for the predator group after the sea urchins were not fed for a week, resulting in eight squares of the HFAs. To create sufficient HFAs condition for sea urchins, we subsequently put four M. nudus in the tank (Fig. 1G). The four M. nudus were not exposed to predators in the control group with the same fitness in all squares (Fig. 1H). The predator test was repeated 8 times using different sea urchins and crabs (N = 8). Eight sea urchins were used in each replicate (four for the experimental group, four for the control group).

Statistical analysis

We calculated the total number of squares between sea urchins and the nets, the number of sea urchins in the HFAs and the number of aggregated sea urchins in groups according to the 30 min video recordings. The total number of squares between sea urchins and nets was the sum of the minimum distance (measured in squares) between each sea urchin and the net. Aggregation was considered as the grouping of two or more sea urchins, according to the definition of aggregation (Vadas et al., 1986). The corresponding squares to the HFAs in the food/predator group were used as the HFAs in the control group.

The data were analyzed for homogeneity of variance and normal distribution using Levene’s test and Kolmogorov–Smirnov test before statistical analysis, respectively. All data showed normal distribution and homogeneity of variance in Experiment 1, while some of the data in Experiment 2 did not show normal distribution and/or homogeneity of variance. In Experiment 1, therefore, independent sample t-tests were performed to detect differences in the number of squares between sea urchins and the nets, the number of aggregated sea urchins and the number of sea urchins in HFAs. In Experiment 2, independent sample t-tests were performed to detect differences in the number of squares between sea urchins and the nets, the number of aggregated sea urchins in food test. The nonparametric Mann–Whitney test was used to analyze the number of sea urchins in HFAs in the food test, the number of aggregated sea urchins and the number of sea urchins in HFAs in the predator test. All statistical analyses were performed using SPSS 25.0 statistical software. A probability level of P <0.05 was considered as being significant.

Experiment 1

In the food test, the number of aggregated sea urchins was significantly larger in the food group (10.50 ± 2.39) than in the control group (4.38 ± 3.16, df = 7, F = 0.962, P = 0.009, Fig. 2A). The number of squares between sea urchins and the nets was significantly smaller in the experimental group (11.00 ± 4.63 squares) than that in the control group (18.25 ± 5.38, df = 7, F = 0.024, P = 0.012, Fig. 2B). The number of sea urchins in the HFAs of the food group (8.13 ± 1.56) was significantly larger than that of the control group (3.50 ± 1.93, df = 7, F = 0.057, P < 0.001, Fig. 2C).

In the predator test, the number of aggregated sea urchins of the predator group (6.13 ± 1.81) was significantly larger than that of the control group (3.88 ± 1.51, df = 7, F = 0.047, P = 0.009, Fig. 2D). The number of squares between sea urchins and the nets was significantly larger in the predator group (17.88 ± 4.02) than that in the control group (11.50 ± 5.61, df = 7, F = 1.757, P = 0.020, Fig. 2E). The number of sea urchins in the HFAs was significantly larger in the predator group (5.25 ± 1.28) than those in the control group (2.88 ± 1.64, df = 7, P = 0.006, Fig. 2F).

Experiment 2

In the food test, no significant difference was found in the number of aggregated sea urchins between the control and the food groups (df = 7, F = 0.364, P = 0.103, Fig. 3A). The number of squares between sea urchins and the nets was significantly larger in the control group (9.25 ± 3.01 squares) than that in the food group (3.88 ± 3.76 squares, df = 7, F = 0.559, P = 0.007, Fig. 3B). The number of sea urchins in the HFAs was significantly larger in the food group (4.25 ± 1.67) than that in the control group (1.88 ± 1.36, df = 7, U = 8.50, P = 0.010, Fig. 3C).

In the predator test, no significant difference of aggregation was found between the control and the predator group (df = 7, U = 26.50, P = 0.574, Fig. 3D). The number of squares between sea urchins and the nets was significantly bigger in the predator group (9.50 ± 2.39) than that in the control group (5.75 ± 1.49, df = 7, F = 3.402, P = 0.002, Fig. 3E). The number of sea urchins in the HFAs was significantly larger in the predator group (2.88 ± 0.99) than in the control group (1.63 ± 0.92, df = 7, U = 12.50, P = 0.038, Fig. 3F).