Scottish Marine and Freshwater Science Volume 4 Number 2: Connectivity of benthic priority marine species within the Scottish MPA network

Results

Cnidaria

Tall Sea Pen ( Funiculina quadrangularis)

As tall sea pen spawn between October and January and the PLD and settlement competency period of a similar species is 65 days, in our connectivity simulations tall sea pens were assumed to spawn in autumn and winter, with a settlement window of 28-65 days. Based on presence data, particles were released from the following MPAs: SMI, NWS, CFL, CFL, SSH, GSH and BHT.

Figure 3a shows the distribution of particles at the end of the settlement window (right panels) and identifies the MPAs locations that these particles drifted over during that period (left panels) for each of the spawning periods (top: autumn; bottom: winter). The potential widespread distribution of offspring and the significant connectivity potential is the result of its relatively long pelagic larval duration ( PLD) period. An index of the distance covered by the larvae in their PLD is presented in Table 3.

Figure 3a: Black dots show the distribution of particles at the end of the settlement window (right panels) and MPAs locations that these particles drifted over during that period (left panels) for each of the spawning periods (top: autumn; bottom: winter). Red dots show the particle origin positions.

Table 3

Mean and standard deviation of the distance (straight line) between start and end position, for each particle representing tall sea pen larvae released from each origin MPA, for each spawning season (n is the number of particles released).

MPA	Autumn		Winter		n
	mean	stdev	mean	stdev
SMI	398.7342	186.2706	472.1786	245.7127	2400
NWS	541.8706	155.5513	664.4572	138.8007	1200
CFL	161.4177	109.0815	257.1189	209.5545	2900
CFL	247.5095	208.9594	475.5425	265.7354	1000
SSH	842.7279	94.4515	837.1288	97.0671	8300
GSH	905.5469	88.51035	959.7132	111.8897	9100
BHT	735.0193	130.9657	801.8007	101.1855	18700

The results on Table 3 show considerable variability between MPAs (e.g. those from Geikie Slide and Hebridean Slope ( GSH) cover considerably longer distance than those from Central Fladen ( CFL). Particles released in winter tend to cover longer distances but display greater variability, compared to those in the autumn.

Figure 3b shows a connectivity matrix between origin and (potential) destination MPAs, confirming the pattern that, in the case of tall sea pens, west coast MPAs are considerably more dispersive than North Sea ones.

Figure 3b: Matrix showing the percentage of all particles representing tall sea pen released from each origin MPA drifting over any MPA during the settlement period of their pelagic phase. Grey boxes indicate zero hits.

Northern Sea Fan ( Swiftia pallida)

As the plannular larvae of the northern sea fan, Swiftia pallida, are thought to be have a short pelagic larval duration we assumed a settlement window of one to ten days with spawning in summer and winter. Based on presence data, particles were released from the following MPAs: SMI, SEB, STM, LSW, TBB and FOF.

Figure 4a shows the distribution of particles at the end of the settlement window (right panels) and identifies the MPAs locations that these particles drifted over during that period (left panels) for each of the spawning periods (top: summer; bottom: winter). The potential relatively narrow distribution of offspring and the reduced connectivity potential, including a considerable degree of self-recruitment (see also Figure 4b), is the result of its relatively short PLD period. An index of the distance covered by the larvae in their PLD is presented in Table 4.

Figure 4a: Black dots show the distribution of particles at the end of the settlement window (right panels) and MPAs locations that these particles drifted over during that period (left panels) for each of the spawning periods (top: summer; bottom: winter). Red dots show the particle origin positions.

Table 4

Mean and standard deviation of the distance (straight line) between start and end position, for each particle representing northern sea fan larvae released from each origin MPA, for each spawning season (n is the number of particles released).

MPA	Summer		Winter		n
	mean	stdev	mean	stdev
SMI	58.93132	33.395	223.0987	136.1454	2400
SEB	83.58833	22.53601	388.1834	63.63042	1200
STM	72.56669	28.09435	237.0058	121.064	14100
LSW	21.96786	22.63312	72.0195	37.45215	100
TBB	64.69743	16.32645	135.9349	28.97287	800
FOF	32.82873	15.25553	112.5105	45.14544	8300

The results on Table 4 show quite a lot of variability between MPAs (e.g. those from Loch Sween ( LSW) cover considerably shorter average distances than those from Shiant East Bank ( SEB)). Particles released in winter tend to cover considerably longer distances but display greater variability in general, compared to those in the summer.

Figure 4b shows a connectivity matrix between origin and (potential) destination MPAs, confirming the pattern that, in the case of northern sea fans, west coast MPAs are considerably more dispersive than North Sea ones but, overall, the dispersal potential of this species is limited.

Figure 4b: Matrix showing the percentage of all particles representing northern sea fan released from each origin MPA drifting over any MPA during the settlement period of their pelagic phase. Grey boxes indicate zero hits.

Pink Soft Coral ( Alcyonium hibernicum)

Pink soft coral and pink sea fingers spawn in late summer between August and September and the plannular larvae settle shortly after release. So, for the purpose of our connectivity simulations, pink soft corals were assumed to spawn in summer, with a settlement window of one to ten days. Based on presence data, particles were released from the following MPAs: CSS, STM and LSW.

Figure 5a shows the distribution of particles at the end of the settlement window (right panel) and identifies the MPAs locations that these particles drifted over during that period (left panel). The potential distribution of offspring is quite limited, resulting from a short PLD period and limited distribution at origin within the proposed MPAs. An index of the distance covered by the larvae in their PLD is presented in Table 5.

Figure 5a: Black dots show the distribution of particles at the end of the settlement window (right panel) and MPAs locations that these particles drifted over during that period (left panel). Red dots show the particle origin positions.

Table 5

Mean and standard deviation of the distance (straight line) between start and end position, for each particle representing pink soft coral larvae released from each origin MPA (n is the number of particles released).

MPA	Summer		n
	mean	stdev
CSS	41.40528	27.13667	2800
STM	68.07873	27.38209	14100
LSW	16.73283	17.34437	100

The results on Table 5 show some differences between MPAs, which reflect their location (the closer inshore, the less dispersive).

Figure 5b shows a connectivity matrix between origin and (potential) destination MPAs. There is a considerable degree of self-recruitment, as a result of relatively inshore MPA locations and short PLD.

Figure 5b: Matrix showing the percentage of all particles representing pink soft coral released from each origin MPA drifting over any MPA during the settlement period of their pelagic phase. Grey boxes indicate zero hits.

Burrowing Sea Anemone ( Arachnanthus sars)

The larvae of the Burrowing anemone ( Arachnanthus sarsi) are present in the plankton from April to the autumn so we assumed spawning was in spring, summer and autumn. As the PLD of the related Cerianthus species range from four weeks to four months we used a settlement window of 28-90 days. Based on presence data, particles were released from the following MPAs: SMI, NWS, STM, EPL, CFL, NWO, SEF, WFL, NSP, CFL, SSH, GSH and BHT.

Figure 6a shows the distribution of particles at the end of the settlement window (three right panels) and identifies the MPA locations that these particles drifted over during that period (three left panels) for each of the three spawning periods (spring to autumn). The potential distribution of offspring and between- MPA connectivity are extremely wide, as a result of a relatively long PLD period and widespread distribution at origin within the proposed MPAs. An index of the distance covered by the larvae in their PLD is presented in Table 6.

Figure 6a: Black dots show the distribution of particles at the end of the settlement window (three right panels) and MPAs locations that these particles drifted over during that period (three left panels) for each of the spawning periods (top: spring and autumn; bottom: summer). Red dots show the particle origin positions.

Table 6

Mean and standard deviation of the distance (straight line) between start and end position, for each particle representing burrowing sea anemone larvae released from each origin MPA, for each spawning season (n is the number of particles released).

MPA	spring		summer		autumn		n
	mean	stdev	mean	stdev	mean	stdev
SMI	262.7722	99.05911	255.1555	123.6793	376.8256	131.2515	2400
NWS	296.6722	92.02234	330.6136	89.24955	465.8216	111.2949	1200
STM	288.2795	90.58127	284.4247	111.7464	385.4184	120.9899	14100
EPL	431.3364	98.19427	470.2425	91.2673	566.0165	127.0553	2200
CFL	164.8757	128.6385	178.141	133.9859	202.9292	133.3907	2900
NWO	497.0089	122.4901	501.516	106.2314	537.7676	118.4944	17500
SEF	317.3321	86.67873	385.8749	96.25372	390.7663	102.9626	1700
WFL	268.6549	163.4696	273.0761	141.7989	346.1803	118.9167	3000
NSP	360.5706	86.3769	399.9785	109.7099	427.3958	95.44053	800
CFL	175.232	143.4311	184.7308	140.4058	220.932	134.978	1000
SSH	520.0574	76.05066	538.3254	80.94968	543.814	83.85678	8300
GSH	575.8729	73.28264	584.3992	72.81853	599.6261	78.29477	9100
BHT	504.5871	122.0041	544.7715	108.1969	615.5512	69.10734	18700

The results on Table 6 confirm the general patterns observed for other species, i.e. that offshore MPAs tend to be more dispersive than inshore ones, and offspring spawned in west coast MPAs tend to cover greater distances than those in the North Sea. Autumn spawned larvae also tend to travel further than those spawned earlier in the year.

Figure 6b shows a connectivity matrix between origin and (potential) destination MPAs. The results are consistent with the patterns described above (Figure 6a and Table 6).

Figure 6b: Matrix showing the percentage of all particles representing burrowing sea anemone released from each origin MPA drifting over any MPA during the settlement period of their pelagic phase. Grey boxes indicate zero hits.

Mollusca

Horse Mussel ( Modiolus modiolus)

The bivalve Modiolus modiolus has peaks of spawning in spring and early summer and the planktonic veliger stage duration was found to take approximately 38 days. So, for the purpose of our connectivity simulations, horse mussels were assumed to spawn all year-round, with a settlement window of 30-40 days. Based on presence data, particles were released from the following MPAs: NOH, FTH and SMI.

Figure 7a shows the distribution of particles at the end of the settlement window (right panels) and identifies the MPAs locations that these particles drifted over during that period (left panels) for each of the spawning periods (rows from the top: spring, summer, autumn and winter, respectively). The potential distribution of offspring is relatively wide but connectivity potential is not very strong (see Figure 7b too), resulting from a relatively long PLD period but limited distribution at origin within the proposed MPAs. An index of the distance covered by the larvae in their PLD is presented in Table 7.

Figure 7a: Black dots show the distribution of particles at the end of the settlement window (right panels) and MPAs locations that these particles drifted over during that period (left panels) for each of the spawning periods (rows from the top: spring to winter). Red dots show the particle origin positions.

Table 7

Mean and standard deviation of the distance (straight line) between start and end position, for each particle representing horse mussel larvae released from each origin MPA, for each spawning season (n is the number of particles released).

MPA	spring		summer		autumn		winter		n
	mean	stdev	mean	stdev	mean	stdev	mean	stdev
NOH	249.0141	83.19418	232.8721	84.93978	260.3104	78.66676	281.9295	77.72447	100
FTH	213.35	60.20469	221.2325	69.44247	216.3795	54.13975	220.4248	43.37147	600
SMI	178.4879	106.1896	176.5887	109.0172	246.9048	131.9025	292.0999	165.5732	2400

The results on Table 7 show relatively small differences between MPAs and seasons, compared to other species.

Figure 7b shows a connectivity matrix between origin and (potential) destination MPAs. As with other species, the west coast MPA is more dispersive than North Sea ones although this could be partly due to the distribution of MPAs along the west coast, since the dispersal distances were not very different between areas.

Figure 7b: Matrix showing the percentage of all particles representing horse mussel released from each origin MPA drifting over any MPA during the settlement period of their pelagic phase. Grey boxes indicate zero hits.

Ocean Quahog ( Arctica islandica)

As spawning in the ocean quahog is protracted and the duration of the larval phase is approximately 32-55 days we assumed spawning was in summer and autumn, with a settlement window of 40-60 days for the purpose of our connectivity simulations. Based on presence data, particles were released from the following MPAs: NWS, STR, ARR, GLE, EGM, WFL, NSP, BHT, FSS and FOF.

Figure 8a shows the distribution of particles at the end of the settlement window (right panels) and identifies the MPAs locations that these particles drifted over during that period (left panels) for each of the spawning periods (top: spring; bottom: summer). The potential distribution of offspring is very wide and connectivity potential is considerable (see also Figure 8b), as a result of the long PLD period and extensive distribution at origin within the proposed MPAs. An index of the distance covered by the larvae in their PLD is presented in Table 8.

Figure 8a: Black dots show the distribution of particles at the end of the settlement window (right panels) and MPAs locations that these particles drifted over during that period (left panels) for each of the spawning periods (top: summer; bottom: autumn). Red dots show the particle origin positions.

Table 8

Mean and standard deviation of the distance (straight line) between start and end position, for each particle representing ocean quahog larvae released from each origin MPA, for each spawning season (n is the number of particles released).

MPA	Summer		Autumn		n
	mean	stdev	mean	stdev
NWS	210.3894	71.44167	301.467	87.5921	1200
STR	197.8751	71.33853	212.3195	44.46416	7600
ARR	75.12121	62.87876	50.64271	51.43547	1300
GLE	202.6309	55.94869	280.0399	80.62585	300
EGM	304.72	114.9622	221.5995	96.61362	3800
WFL	163.6299	106.9408	254.4092	114.8778	3000
NSP	336.094	139.4325	292.0905	125.6298	800
BHT	349.156	131.5106	471.0206	80.57485	18700
FSS	416.9898	81.69673	397.0485	94.58714	25500
FOF	110.676	53.35329	113.9773	52.88048	8300

The results on Table 8 show wide differences between MPAs (e.g. considerable distances covered by larvae originating from offshore MPAs such as the Faroe-Shetland Sponge Belt ( FSS) but smaller in the case of more coastal MPA, with Arran ( ARR) as an extreme example. The differences between seasons are not very large, nor are there consistent patterns between MPAs.

Figure 8b shows a connectivity matrix between origin and potential destination MPAs. As with other species, the west coast MPA is more dispersive than North Sea ones although this could be partly due to the distribution of MPAs along the west coast. The patterns observed for dispersal distance were not particularly obvious in the connectivity matrix, probably masked by the east-west coast differences mentioned above.

Figure 8b: Matrix showing the percentage of all particles representing ocean quahog released from each origin MPA drifting over any MPA during the settlement period of their pelagic phase. Grey boxes indicate zero hits.

Fan Mussel ( Atrina fragilis)

Fan mussels have been reported to spawn in the summer so we assumed spawning was from spring to autumn, with a settlement window of 30-50 days. Based on presence data, particles were released from the following MPAs: SMI, CSS and ARR.

Figure 9a shows the distribution of particles at the end of the settlement window (right three panels) and identifies the MPAs locations that these particles drifted over during that period (left three panels) for each of the three spawning periods (spring, autumn and winter). The potential distribution of offspring is quite wide, with origin MPAs in the west coast potentially being able to contribute offspring to areas in the north and the east, as a result of a relatively long PLD period. An index of the distance covered by the larvae in their PLD is presented in Table 9.

Figure 9a: Black dots show the distribution of particles at the end of the settlement window (three right panels) and MPAs locations that these particles drifted over during that period (three left panels) for each of the spawning periods (top: spring and autumn; bottom: summer). Red dots show the particle origin positions.

Table 9

Mean and standard deviation of the distance (straight line) between start and end position, for each particle representing fan mussel larvae released from each origin MPA, for each spawning season (n is the number of particles released).

MPA	spring		summer		autumn		n
	mean	stdev	mean	stdev	mean	stdev
SMI	215.0326	122.1579	215.8706	132.1664	303.2008	150.6005	2400
CSS	133.2396	100.5229	103.1948	93.99843	89.98832	88.15232	2800
ARR	110.5772	84.0085	82.20794	62.74385	55.20894	50.6603	1300

The results on Table 9 show a range of average distances that reflects the inshore-offshore gradient of origin. The results are quite variable and there are seasonal differences without a clear pattern between MPAs.

Figure 9b shows a connectivity matrix between origin and (potential) destination MPAs. The results are consistent with the patterns observed in Figure 9a. The matrix also shows, as expected, a greater degree of self-recruitment within Clyde Sea MPAs.

Figure 9b: Matrix showing the percentage of all particles representing fan mussel released from each origin MPA drifting over any MPA during the settlement period of their pelagic phase. Grey boxes indicate zero hits.

Native Oyster ( Ostrea edulis)

As the larvae of the native oyster are pelagic for 11-30 days, we used a settlement window of 10-30 days, with spawning in the summer. Based on presence data, particles were released from the following MPAs: NWS, STM, LSW, GLE and WYR.

Figure 10a shows the distribution of particles at the end of the settlement window (right panel) and identifies the MPAs locations that these particles drifted over during that period (left panel). Even though the origin MPAs are all on the west coast, the potential distribution of offspring is relatively wide, resulting from a relatively long PLD period. An index of the distance covered by the larvae in their PLD is presented in Table 10.

Figure 10a: Black dots show the distribution of particles at the end of the settlement window (right panel) and MPAs locations that these particles drifted over during that period (left panel). Red dots show the particle origin positions.

Table 10

Mean and standard deviation of the distance (straight line) between start and end position, for each particle representing native oyster larvae released from each origin MPA, for each spawning season (n is the number of particles released).

MPA	summer		n
	mean	stdev
NWS	163.35	78.06062	1200
STM	162.339	73.02932	14100
LSW	62.05346	41.44984	100
GLE	151.0637	67.76181	300
WYR	244.9038	89.90069	100

The results on Table 10 show relatively small differences between MPAs, considering that some of these were very small areas, where only small numbers of particles were released (one to three release positions for three of the origin MPAs).

Figure 10b shows a connectivity matrix between origin and (potential) destination MPAs. The results show the potential for relatively long distance transport. Relatively large connectivity values may have been influenced by the small size of some origin MPAs and, consequently, the small number of particles released from those.

Figure 10b: Matrix showing the percentage of all particles representing native oyster released from each origin MPA drifting over any MPA during the settlement period of their pelagic phase. Grey boxes indicate zero hits.

Flame Shell ( Limaria hians)

Flame shell spawn from May-June with peak settlement from July-August in Scottish waters so, for the purpose of our connectivity simulations, flame shell were assumed to spawn in summer, with a settlement window of 20-60 days. Based on presence data, particles were released only from NWS MPA (North-west sea lochs and Summer Isles).

Figure 11 shows the distribution of particles at the end of the settlement window (right) and identifies the MPAs locations that these particles drifted over during that period (left). The potential distribution of offspring is relatively wide, considering the restricted origin, as offspring can potentially reach northern and eastern MPAs. An index of the distance covered by the larvae in their PLD is presented in Table 11a.

Figure 11: Black dots show the distribution of particles at the end of the settlement window (right panel) and MPAs locations that these particles drifted over during that period (left panel). Red dots show the particle origin positions.

Table 11a

Mean and standard deviation of the distance (straight line) between start and end position, for each particle representing flame shell larvae released from the origin MPA (n is the number of particles released).

MPA	summer		n
	mean	stdev
NWS	210.3894	71.44167	1200

Table 11a is not very informative because only one MPA and season were considered, and the distance between the origin and end-point of the particles ignore the fact that many travelled a considerably longer distance, around the northern coast of Scotland (mainland and northern isles, as it can be inferred from Figure 11).

Instead of a highly one-dimensional connectivity matrix between the single origin MPA and all other (potential) destination MPAs, the percentage of all particles representing flame shell larvae that drift over any MPA during the settlement period of their pelagic phase is presented in Table 11b. The results confirm the patterns observed in Figure 11.

Table 11b

Percentage of all particles representing flame shell larvae released from the origin MPA that drift over any MPA during the settlement period of their pelagic phase.

	Destination
Origin	NOH	NWS	STR	ECC	PWY	WYR	TBB	NWO	FOF
NWS	1.083	0.667	37.25	2.833	1.75	0.667	5.417	30.58	1.5

Heart Cockle ( Glossus humanus)

In the Clyde, heart cockle probably spawn at the end of September and their larvae tend to be in the plankton for a few weeks. Therefore, we assumed that heart cockles spawn in autumn, with a settlement window of 1-30 days. Based on presence data, particles were released from a single MPA: GLE.

Figure 12 shows the distribution of particles at the end of the settlement window (right panel) and identifies the MPAs locations that these particles drifted over during that period (left panel). The potential distribution of offspring is relatively wide but disperse, resulting from a moderately long PLD period but very limited distribution at origin within the proposed MPAs (only one small MPA in the simulations, only three distinct particle release positions). An index of the distance covered by the larvae in their PLD is presented in Table 12a.

Figure 12: Black dots show the distribution of particles at the end of the settlement window (right panel) and MPAs locations that these particles drifted over during that period (left panel). Red dots show the particle origin positions.

Table 12a

Mean and standard deviation of the distance (straight line) between start and end position, for each particle representing heart cockle larvae released from the origin MPA (n is the number of particles released).

MPA	autumn		n
	mean	stdev
GLE	157.583	47.71967	300

Instead of a highly one-dimensional connectivity matrix between the single origin MPA and all other (potential) destination MPAs, the percentage of all particles representing heart cockle larvae that drift over any MPA during the settlement period of their pelagic phase is presented in Table 12b. The results confirm the patterns observed in Fig. 12.

Table 12b

Percentage of all particles representing heart cockle larvae released from the origin MPA that drift over any MPA during the settlement period of their pelagic phase.

	Destination
Origin	NWS	STR	SEB	LSW	GLE	NWO
GLE	87.00	0.33	0.67	14.00	15.67	12.00

Most of the heart cockle larvae are transported along the coast to MPA NWS, although other MPAs in the vicinity are also potentially visited and there is some self-recruitment. The offspring can potentially reach MPAs in Orkney and into the northern North Sea.

Arthropoda

Amphipod (Maera loveni)

As amphipods have no larval stages and dispersion is limited to crawling, swimming, and "rafting" on algae, we assumed these amphipods spawned all year-round, with a settlement window of one to ten days. Based on presence data, particles were released from the following MPAs: NWS, STR, ARR, GLE, EGM, WFL, NSP, BHT, FSS and FOF.

Figure 13a shows the distribution of particles at the end of the settlement window (right panels) and identifies the MPAs locations that these particles drifted over during that period (left panels) for each of the spawning periods (top to bottom: spring to winter). The potential distribution of offspring is very limited, as a result of the life history characteristics of this animal. An index of the distance covered by the offspring is presented in Table 13.

Figure 13a: Black dots show the distribution of particles at the end of the settlement window (right panels) and MPAs locations that these particles drifted over during that period (left panels) for each of the spawning periods (rows from the top: spring to winter). Red dots show the particle origin positions.

Table 13

Mean and standard deviation of the distance (straight line) between start and end position, for each particle representing amphipod offspring released from each origin MPA, for each spawning season (n is the number of particles released).

MPA	spring		summer		autumn		winter		n
	mean	stdev	mean	stdev	mean	stdev	mean	stdev
NWS	45.71541	19.9395	40.61786	18.85419	49.44458	20.30217	60.3965	23.25147	1200
STR	73.30068	17.13326	72.94866	17.12485	74.44586	18.12632	72.37808	19.24067	7600
ARR	29.70384	16.22294	29.56981	16.26043	28.08696	14.25608	27.70105	14.30954	1300
GLE	50.26104	18.6439	43.82027	16.52504	52.9068	19.21399	64.03978	21.8383	300
EGM	35.99597	11.65453	36.03197	11.9373	32.72312	10.17337	32.60442	9.312925	3800
WFL	52.03145	14.06442	45.3174	13.63996	54.99862	15.19404	55.52642	14.71882	3000
NSP	50.32005	7.560243	48.08747	7.108671	51.16166	7.499674	52.05183	7.24344	800
BHT	33.40955	16.599	34.53022	18.4953	43.34478	19.14527	48.62432	16.13506	18700
FSS	102.975	18.09356	101.692	17.26555	112.0983	17.61639	114.0385	18.46735	25500
FOF	25.78633	11.99884	27.46618	12.67583	29.98636	13.01473	31.95397	13.57806	8302

The results on Table 13 show relatively small differences between MPAs and seasons, compared to other species. The only origin MPA with greater dispersal potential than the rest is the Faroe-Shetland Sponge Belt ( FSS), due to its hydrographic characteristics.

Figure 13b shows a connectivity matrix between origin and (potential) destination MPAs. Most origin MPAs are quite retentive, given the biological characteristics of this species, with a small number of relative exceptions.

Figure 13b: Matrix showing the percentage of all particles representing amphipods released from each origin MPA drifting over any MPA during the settlement period of their pelagic phase. Grey boxes indicate zero hits.

Spiny Lobster ( Palinurus elephas)

For the purpose of our connectivity simulations, spiny lobsters were assumed to spawn in summer as they produce a clutch from around July to October. Larvae may drift for one to six months so a settlement window of 60-180 days was used in the simulation. Based on presence data, particles were released from the following MPAs: SMI, NWS, STR and LSW.

Figure 14a shows the distribution of particles at the end of the settlement window (right panel) and identifies the MPAs locations that these particles drifted over during that period (left panel). The potential distribution of offspring and connectivity potential are extremely wide, given the long PLD period of this species. An index of the distance covered by the larvae in their PLD is presented in Table 14.

Figure 14a: Black dots show the distribution of particles at the end of the settlement window (right panel) and MPAs locations that these particles drifted over during that period (left panel). Red dots show the particle origin positions.

Table 14

Mean and standard deviation of the distance (straight line) between start and end position, for each particle representing spiny lobster larvae released from each origin MPA (n is the number of particles released).

MPA	summer		n
	mean	stdev
SMI	626.0166	197.1344	2400
NWS	690.7814	118.3799	1200
STR	481.381	202.9256	7600
LSW	139.3821	127.9197	100

The results on Table 14 show some differences between MPAs. The general pattern shows smaller distances closer inshore and more dispersive MPAs in the west, compared to the east coast.

Figure 14b shows a connectivity matrix between origin and (potential) destination MPAs. As with other species, west coast MPAs are more dispersive than North Sea ones. A relatively high proportion of offspring originating from the Loch Sween MPA have the potential to colonise MPAs in the Clyde area (Clyde Sea Sill ( CSS) and Arran ( ARR)).

Figure 14b: Matrix showing the percentage of all particles representing spiny lobster released from each origin MPA drifting over any MPA during the settlement period of their pelagic phase. Grey boxes indicate zero hits.

Echinodermata

Northern Feather Star ( Leptometra celtica)

The free-swimming vitellaria larvae of northern feather star only last a few days before settlement and so a settlement window of one to ten days was assumed. Due to the limited information on spawning times, we assumed that spawning occurred all year-round. Based on presence data, particles were released from the following possible MPAs; SMI, NWS, MTB and LSW.

Figure 15a shows the distribution of particles at the end of the settlement window (right panels) and identifies the MPAs locations that these particles drifted over during that period (left panels) for each of the spawning periods (top to bottom: spring to winter). The potential distribution of offspring is low, resulting from the short PLD period, and quite constant between seasons. An index of the distance covered by the larvae in their PLD is presented in Table 15.

Figure 15a: Black dots show the distribution of particles at the end of the settlement window (right panels) and MPAs locations that these particles drifted over during that period (left panels) for each of the spawning periods (rows from the top: spring to winter). Red dots show the particle origin positions.

Table 15

Mean and standard deviation of the distance (straight line) between start and end position, for each particle representing northern feather star larvae released from each origin MPA, for each spawning season (n is the number of particles released).

MPA	spring		summer		autumn		winter		n
	mean	stdev	mean	stdev	mean	stdev	mean	stdev
SMI	58.80072	34.80895	58.93132	33.395	63.88066	37.24751	74.00236	45.65417	2400
NWS	52.20294	22.69311	46.40062	20.75867	57.20983	24.6535	72.89097	32.6751	1200
MTB	65.67233	26.93485	68.35814	29.93843	75.50844	29.67039	70.20467	26.56152	200
LSW	24.93985	20.92754	21.96786	22.63312	25.41234	25.30455	25.20322	21.74939	100

The results on Table 15 show relatively small differences between MPAs, with the exception of a very inshore one (Loch Sween ( LSW)) and seasons, although winter spawning was generally more dispersive.

Figure 15b shows a connectivity matrix between origin and (potential) destination MPAs. Inshore MPAs like LSW or North-west sea lochs and Summer Isles ( NWS) are more retentive (resulting in more self-recruitment) than further offshore MPAs.

Figure 15b: Matrix showing the percentage of all particles representing northern feather star released from each origin MPA drifting over any MPA during the settlement period of their pelagic phase. Grey boxes indicate zero hits.

Analysis of General Patterns

The GAM fitted to mean transport distance estimates for each species, MPA and season (QUARTER: SPR, SUM, AUT, WIN) combination indicated that maximum pelagic larval duration ( PLDMAX; days) and Distance to shore (Distancetoshore; km) had a significant effect on mean transport distance (Table 16). However, only winter had a significantly greater transport rate than other quarters and OSPAR region was not significant when distance from shore was considered.

Table 16

Outcome of fitting successively more complex models up to the minimum adequate model ( GAM) to mean particle distance travelled.

Model	d.f.	Deviance explained (%)	F-values	P-values
s( PLDMAX)	1	49	28.49	<0.001
+ fQuarter	3	11.5	8.56	<0.001
+ s(Distance to shore)	1	5.6	4.38	0.017