UBC Faculty Research and Publications

An Ecosystem Model of the Ocean Around Haida Gwaii, Northern British Columbia : Ecopath, Ecosim and Ecospace Kumar, Rajeev; Surma, Szymon; Pitcher, Tony J.; Varkey, Divya; Lam, Mimi E.; Ainsworth, Cameron H.; Pakhomov, Evgeny 2016

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
52383-Kumar_R_et_al_Ecosystem_model_ocean_around.pdf [ 19.88MB ]
Metadata
JSON: 52383-1.0354480.json
JSON-LD: 52383-1.0354480-ld.json
RDF/XML (Pretty): 52383-1.0354480-rdf.xml
RDF/JSON: 52383-1.0354480-rdf.json
Turtle: 52383-1.0354480-turtle.txt
N-Triples: 52383-1.0354480-rdf-ntriples.txt
Original Record: 52383-1.0354480-source.json
Full Text
52383-1.0354480-fulltext.txt
Citation
52383-1.0354480.ris

Full Text

ISSN 1198-6727  Fisheries Centre Research Reports 2016   Volume 24   Number 2  An Ecosystem Model of the Ocean Around Haida Gwaii, Northern British Columbia: Ecopath, Ecosim and Ecospace         Institute for the Oceans and Fisheries,  University of British Columbia, Canada  2016  Fisheries Centre Research Reports 24(2) 		ii																																				Kumar,	R.,	Surma,	S.,	Pitcher,	T.J.,	Varkey,	D.,	Lam,	M.,	Ainsworth,	C.	and	Pakhomov,	E.	(2016)		An	Ecosystem	Model	of	the	Ocean	Around	Haida	Gwaii,	Northern	British	Columbia:		Ecopath,	Ecosim	and	Ecospace.	Fisheries	Centre	Research	Reports	24	(2):	76pp.		©		Institute	for	the	Oceans	and	Fisheries,	University	of	British	Columbia	2016	Fisheries	Centre	Research	Reports	are	Open	Access	publications		An Ecosystem Model of the Ocean Around Haida Gwaii, Northern British Columbia: Ecopath, Ecosim and Ecospace 	Rajeev	Kumar1,	Szymon	Surma1,	Tony	J.	Pitcher1,	Divya	Varkey1,	Mimi	Lam1,	Cameron	Ainsworth2	and	Evgeny	Pakhomov1	1Institute	for	the	Oceans	and	Fisheries,	University	of	British	Columbia	2College	of	Marine	Science,	University	of	South	Florida																				©		Institute	for	the	Oceans	and	Fisheries,	University	of	British	Columbia	2016	Fisheries	Centre	Research	Reports	are	Open	Access	publications		Please	Cite	as:	Kumar,	R.,	Surma,	S.,	Pitcher,	T.J.,	Varkey,	D.,	Lam,	M.,	Ainsworth,	C.	and	Pakhomov,	E.	(2016)	An	Ecosystem	Model	of	the	Ocean	Around	Haida	Gwaii,	Northern	British	Columbia:	Ecopath,	Ecosim	and	Ecospace.	Fisheries	Centre	Research	Reports	24	(2):	76pp.		 	2016  Fisheries Centre Research Reports 24(2) 		2	Table of Contents  Table	of	Contents	...........................................................................................................................	2	List	of	Figures	..................................................................................................................................	3	List	of	Tables	...................................................................................................................................	3	Abstract	..........................................................................................................................................	4	Introduction	....................................................................................................................................	5	The	development	of	ecosystem	models	for	British	Columbia	waters	............................................	6	Adapting	the	NBC	EwE	model	for	the	Haida	Gwaii	(HG)	ecosystem	..............................................	8	Methods	.........................................................................................................................................	8	Ecopath	.......................................................................................................................................	8	Ecosim	........................................................................................................................................	9	Ecospace	...................................................................................................................................	10	Designing	the	Ecospace	map	................................................................................................	10	Movement	between	cells	.....................................................................................................	11	Spatial	representation	of	fishing	effort	................................................................................	12	Setting	up	scenarios	.............................................................................................................	12	EwE	version	upgrading	.................................................................................................................	13	Delimitation	of	the	study	area	......................................................................................................	13	Parameterization	of	readjusted	functional	groups	......................................................................	14	Functional	groups	.................................................................................................................	17	Addition	of	new	fisheries	..........................................................................................................	35	Incorporation	of	spatial	information	............................................................................................	36	Delineation	of	the	boundaries	of	the	Haida	Gwaii	Ecospace	map	.......................................	36	Ecospace	habitat	capacity	....................................................................................................	38	Spatial	distribution	of	fishing	fleets	......................................................................................	46	Analyses	of	marine	protected	areas	.....................................................................................	48	PART	B:	Another	improvement	to	the	HG	ecosystem	model	.......................................................	49	Pacific	herring	...........................................................................................................................	49	Parameterization	of	Pacific	herring	......................................................................................	49	New	groups	to	be	introduced	to	the	HG	model	.......................................................................	53	Saury	.....................................................................................................................................	53	Epifaunal	invertebrates	........................................................................................................	54	Sea	urchins	...........................................................................................................................	54	Other	grazers	........................................................................................................................	54	Epifaunal	filter-feeders	.........................................................................................................	54	Octopus	................................................................................................................................	54	Epifaunal	carnivores	.............................................................................................................	55	Macrozooplankton	...............................................................................................................	55	Pelagic	amphipods	................................................................................................................	55	Small	gelatinous	zooplankton	...............................................................................................	55	Microzooplankton	................................................................................................................	55	Eelgrass	.................................................................................................................................	55	Macroalgae	...........................................................................................................................	56	A Revised EwE Model of Haida Gwaii 		 3	Canopy	kelp	..........................................................................................................................	56	Benthic	macroalgae	..............................................................................................................	56	Benthic	microalgae	...............................................................................................................	56	Modifications	to	existing	functional	groups	.............................................................................	56	Humpback	whales	................................................................................................................	56	Minke	whales	.......................................................................................................................	57	References	...................................................................................................................................	58	Appendices	...................................................................................................................................	66	Appendix	A.	Revised	HG	ecosystem	model:	parameterization	................................................	66	Appendix	B.	Revised	HG	model:	diet	matrix	............................................................................	68	Appendix	C:	Revised	HG	model:	fisheries	(kg.km-2)	.................................................................	74	Appendix	D.	R	code	for	extracting	relevant	data	from	ArcGIS	files	for	Ecospace	maps	...........	75		List of Figures Figure	1.	Relationship	between	vulnerability	and	predation	mortality	in	Ecosim.	........................	7	Figure	2.	Habitat-based	foraging	capacity.	...................................................................................	11	Figure	3.	Study	area.	....................................................................................................................	13	Figure	4.	Biomass	spectrum	of	the	Haida	Gwaii	food	web	..........................................................	14	Figure	5.	A	 comparison	of	NBC	ecosystem	model	 and	HG	ecosystem	model	 functional	 groups	and	Ecopath	parameters	..............................................................................................................	17	Figure	6.	Plot:	functional	groups	against	trophic	levels.	..............................................................	17	Figure	7.	Predator	consumption	of	Pacific	herring	in	the	HG	ecosystem	model.	........................	29	Figure	8.	Diets	matrix	used	in	HG	model.	.....................................................................................	34	Figure	9.	Total	landings	for	each	fleet	included	in	the	HG	model	................................................	35	Figure	10.	Total	landings	of	each	functional	group	in	the	model,	categorized	by	fleet.	..............	36	Figure	11.	The	Ecospace	base	map	for	the	HG	model.	.................................................................	37	Figure	12.	Extraction	of	HG	data	for	Ecospace	from	GIS	files.	.....................................................	38	Figure	13.	Habitat	capacity	maps	.................................................................................................	41	Figure	14.	MPA	and	spillover	boundaries	in	Ecospace	based	on	high	clumping	MPA	options.	...	48	Figure	15.	Consumption	of	Pacific	herring	by	all	its	predators	in	the	HG	models.	.......................	53	Figure	16.	Proportions	of	herring,	forage	fish	and	euphausiids	in	humpback	whale	diets		.........	53		List of Tables Table	1.	Shape	files	obtained	from	the	Haida	Ocean	Technical	Team	(HOTT)	.............................	39	Table	2.	The	ArcGIS	files	used	for	mapping	the	spatial	distribution	of	fleets	in	the	model	.........	46	Table	3.	Age-structured	parameterization	of	Pacific	herring	across	the	four	stocks	in	the	revised	HG	model	(year	2000).	.................................................................................................................	51	Table	4.	Prey	composition	(by	weight)	of	the	Pacific	herring	in	the	Central	BC	Coast.	................	52		 	2016  Fisheries Centre Research Reports 24(2) 		4	Abstract 	This	 report	describes	 the	 structure	and	workings	of	 an	Ecopath	with	Ecosim	 (EwE)	ecosystem	model	 for	 the	 waters	 surrounding	 Haida	 Gwaii,	 British	 Columbia.	 The	model	 area	 lies	 in	 the	southernmost	part	of	the	Gulf	of	Alaska	Large	Marine	Ecosystem,	and	hence	 is	predominantly	boreal	 in	 species	 composition,	 although	with	more	 commonalities	with	 the	California	Current	than	are	present	in	more	northern	waters.	The	report	covers	the	development	and	parameters	of	the	static	Ecopath	food	web	model	from	its	early	antecedents	(the	Hecate	Strait	models)	through	the	recent	(2000)	and	historical	(1750,	1900,	 1950)	Northern	 British	 Columbia	models	 to	 its	 current	 form	 and	 future	 improvements.	The	 current	 version	 of	 the	 model	 contains	 78	 functional	 groups	 spanning	 five	 trophic	 levels	(including	 increased	 functional	 group	 resolution	 for	 marine	 mammals,	 elasmobranchs	 and	herring	 age	 classes	 and	 stocks)	 and	 21	 fisheries	 (including	 commercial,	 recreational	 and	aboriginal	fleets).	Future	improvements	to	the	Ecopath	model	include	improved	representation	of	 Pacific	 herring	 (age/size	 classes,	 stock	 structure	 and	 diet	 composition),	 zooplankton	 and	epifaunal	invertebrates.	The	report	also	addresses	the	use	of	this	model	as	a	platform	for	dynamic	simulations	in	Ecosim	(including	the	vulnerability	parameters	based	on	foraging	arena	theory)	and	spatial	analysis	 in	Ecospace.	The	latter	includes	the	sources	and	processes	involved	in	the	designation	of	the	base	map,	 habitat	 capacity	 and	 sailing	 cost	 maps	 for	 functional	 groups	 and	 fisheries,	 and	 marine	protected	areas.		Acknowledgements 	The	authors	built	this	ecosystem	model	of	Haida	Gwaii	as	part	of	a	Strategic	Partnership	Grant	(STPGP	447247	–	13)	funded	by	the	Canadian	Natural	Science	and	Engineering	Research	Council.	“Advancing	Forage	Fish	Science:	Understanding	the	ecosystem	role	of	Pacific	herring	in	coupled	social-ecological	systems”.	The	Council	of	the	Haida	Nation	and	the	Heiltsuk	Tribal	Council	are	formal	partners	 in	 this	 research	and	have	 contributed	 information	on	herring	populations	 for	this	 model.	 Szymon	 Surma	 and	 Dr	 Rajeev	 Kumar	 built	 the	model	 with	 input	 from	 the	 other	authors	and	from	Dr	Villy	Christensen	at	UBC.	The	report	was	drafted	mainly	by	Rajeev	Kumar	with	 substantial	 editing	 by	 Szymon	 Surma	 and	 by	 the	 other	 authors.	 Ecosystem	 modelling	comprises	Module	 3	 of	 the	NSERC	project	while	 other	 components	 of	 the	project	 are	 run	by	Simon	Fraser	University	(DNA	Analysis	–	Dr	D.	Yang;	Social-Ecological	Analysis	–	Dr	A.	Salomon	)	and	the	University	of	Toronto	(Herring	Population	Dynamics	–	Dr	M.	Krkosek).			 	A Revised EwE Model of Haida Gwaii 		 5	Introduction  The	northeast	Pacific	Ocean	off	British	Columbia	 (BC)	hosts	numerous	 resident	and	migratory	species	 spanning,	 at	 least,	 five	 trophic	 levels	 and	 all	 size	 categories	 from	 phytoplankton	 to	whales.	Many	of	these	species	support	commercial,	recreational	and	(or)	aboriginal	fisheries.	The	Haida	Gwaii	area	lies	in	the	northernmost	part	of	Canada’s	Exclusive	Economic	Zone	(EEZ)	in	the	Pacific	Ocean.	Due	to	its	position	in	the	southeasternmost	corner	of	the	Gulf	of	Alaska	Large	Marine	 Ecosystem	 (LME),	 the	 local	 food	 web	 is	 predominantly	 boreal	 regarding	 species	composition	 but	 displays	 more	 commonalities	 with	 the	 neighbouring	 California	 Current	 LME	than	are	present	in	more	northern	waters.	These	include	greater	abundance	and	more	frequent	incursions	 of	 southern	 species	 such	 as	 Pacific	 hake	 (Merluccius	 productus),	 Pacific	 sardine	(Sardinops	 sagax)	 and	 California	 sea	 lion	 (Zalophus	 californianus),	 as	 well	 as	 a	 slightly	 lower	abundance	of	northern	species	such	as	capelin	(Mallotus	villosus)	and	walleye	pollock	(Theragra	chalcogramma).	Haida	Gwaii	boasts	a	highly	productive	and	diverse	marine	ecosystem,	including	kelp	forests	and	eelgrass	 beds,	 glass	 sponge	 reefs,	 estuaries	 and	 fjords,	 shallow	 reefs	 and	 banks,	 and	 a	continental	shelf	break	located	closer	to	shore	than	anywhere	else	in	the	Northeast	Pacific	east	of	the	Aleutian	Trench.	The	high	productivity	of	this	ecosystem	is	largely	due	to	oceanographic	features	such	as	the	Haida	Eddies	(created	by	the	splitting	of	the	westerly	North	Pacific	Current	into	northerly	and	southerly	branches	near	the	southwest	coast	of	the	islands)	and	the	Hecate	Strait	Front	(generated	by	the	meeting	of	opposing	tidal	flows	from	Dixon	Entrance	and	Queen	Charlotte	Sound).	The	ecological	diversity	of	the	ocean	surrounding	Haida	Gwaii	results	mainly	from	 the	 variety	 of	 bathymetric	 features	 produced	 by	 a	 combination	 of	 continuing	 tectonic	processes,	Pleistocene	glaciation	and	Holocene	sea	level	rise	and	isostatic	rebound	(Barrie	et	al.	2005).	The	marine	ecosystem	surrounding	Haida	Gwaii	has	existed	 in	approximately	 its	present	 form	since	 at	 least	 ~9,500	 BP,	when	 it	 is	 first	 known	 to	 have	 been	 exploited	 by	 aboriginal	 people	(Fedje	 et	 al.	 2005a)	 and	 most	 likely	 since	 ~11,500	 BP,	 i.e.	 the	 end	 of	 the	 Younger	 Dryas	postglacial	 cold	 period	 (Wigen	 2005).	 The	 contents	 of	 the	 ~9,500	 BP	 Kilgii	 Gwaay	 site	 in	southern	 Haida	 Gwaii	 already	 suggests	 the	 presence	 of	 a	 developed	 maritime	 adaptation	(advanced	 fishing	 techniques	 and	heavy	 reliance	on	marine	protein,	 including	 fish,	mammals,	and	 birds)	 among	 its	 inhabitants	 (Fedje	 et	 al.	 2005a).	 Pacific	 herring	 (Clupea	 pallasii)	 was	harvested	 in	 large	quantities	by	~8,230	BP	at	the	Lyell	Bay	South	site	 in	southern	Haida	Gwaii	(Fedje	 et	 al.	 2005b)	 and	 shellfish	 by	 ~5000	 BP	 at	 the	 Cohoe	 Creek	 site	 on	 Masset	 Inlet	(Christensen	&	 Stafford	 2005).	 The	 fully	 developed	Northwest	 Coast	 culture	 pattern	with	 the	intensive	harvesting	of	Pacific	salmon	(Oncorhynchus	spp.)	was	present	on	the	islands	by	~2000	BP	 (Fedje	 &	 Mackie	 2005),	 or	 perhaps	 as	 late	 as	 1,200	 BP,	 substantially	 later	 than	 on	 the	mainland.	Despite	the	increased	reliance	on	salmon,	shellfish,	herring,	rockfish	(Sebastes	spp.),	Pacific	 halibut	 (Hippoglossus	 stenolepis)	 and	 pinnipeds	 continued	 to	 be	 important	 resources	(Orchard	2011).	Trophic	levels	calculated	for	various	marine	vertebrates	based	on	stable	isotope	data	derived	from	skeletal	remains	dating	to	2000-100	BP	(Szpak	et	al.	2009)	indicate	that	the	local	marine	 food	web	did	not	undergo	a	massive	 structural	 change	 in	 the	 last	2000	years,	 in	spite	of	the	overexploitation	of	many	fish	and	mammal	populations.	2016  Fisheries Centre Research Reports 24(2) 		6	Ecosystems	function	in	a	state	of	perpetual	change.	Most	of	these	changes	are	arguably	due	to	climatic	 variation	 and/or	 human	 exploitation	 (Botsford	 et	 al.	 1997;	 Cury	 et	 al.	 2008).	 Human	intervention	 and	 climate	 change	 have	 affected	 the	 biodiversity,	 ecosystem	 structure,	 and	services	of	the	North	Pacific	Ocean,	as	is	evident	in	the	fluctuations	in	the	species	composition,	distribution	and	commercial	 landings	of	several	pelagic	and	groundfish	fisheries	(McFarlane	et	al.	 2000).	 Further	 examples	 include	 the	 reappearance	of	 Pacific	 sardine	 (Sardinops	 sagax)	 off	the	BC	coast	 in	 the	early	1990s	after	 the	complete	collapse	of	 the	California	 stock	 in	 the	 late	1940s	 (McFarlane	 &	 Beamish	 1999),	 northward	 range	 expansion	 of	 migratory	 Pacific	 hake	(Merluccius	productus)	stocks	into	Haida	Gwaii	waters,	variation	in	the	marine	survival	rates	of	Pacific	 salmon	 (Oncorhynchus	 spp.),	 the	 recruitment	 patterns	 of	 many	 groundfish	 species	(McFarlane	et	al.	2000),	and	an	overall	decline	of	nearly	50%	in	BC	commercial	fisheries	landings	from	~0.3	Mt	to	0.13	Mt	since	1990	(Statistics	DFO-Ottawa	Ontario)	(DFO	2015b).	There	 is	 also	 evidence	 of	 recovery	 in	 many	 formerly	 depleted	 marine	 mammal	 populations,	including	 humpback	 (Megaptera	 novaeangliae)	 and	 fin	 whales	 (Balaenoptera	 physalus).	 The	humpback	whale	population	has	recently	maintained	an	annual	growth	rate	of	over	0.04	yr-1	off	the	 west	 coast	 of	 Canada	 (DFO	 2009b).	 It	 has	 been	 hypothesized	 that	 this	 trend	 might	 be	impeding	the	recovery	of	Pacific	herring	(Ford	et	al.	2009),	a	 leading	forage	fish	species	which	shows	no	 clear	 sign	 of	 recovery	 in	many	 regions	 (stock	 areas)	 off	 the	BC	 coast	 despite	 being	closed	 to	 commercial	 fishing	 for	 the	 last	 ten	 years	 (Schweigert	 et	 al.	 2010).	 However,	 our	analysis	 using	 surplus	 production	 models	 and	 ecosystem	 modelling	 (Surma	 &	 Pitcher	 2015)	suggests	 that	humpback	whale	predation	 is	only	one	of	many	 interacting	 factors	affecting	the	status	of	Pacific	herring	stocks	 in	BC	and	Alaska.	Harbour	seal	 (Phoca	vitulina),	Steller	sea	 lion	(Eumetopias	jubatus)	and	gray	whale	(Eschrichtius	robustus)	populations	in	BC	are	considered	to	be	near	their	carrying	capacities	after	having	recovered	from	historical	overhunting	and	culling.	On	the	other	hand,	blue	(B.	musculus)	and	sei	(B.	borealis)	whale	populations	do	not	appear	to	be	recovering	from	historical	depletion,	while	sperm	whale	(Physeter	macrocephalus)	recovery	is	slow	(Surma	&	Pitcher	2015).	The development of ecosystem models for British Columbia waters To	assess	the	ecosystem-wide	impacts	of	fishing	and	climatic	variations	in	the	coastal	waters	off	BC,	 an	 ecosystem	model	 using	 Ecopath	 (Christensen	 &	 Pauly	 1992;	Walters	 et	 al.	 1997)	 was	developed	 in	 late	 1996	 in	 a	 workshop	 organized	 at	 the	 UBC	 Fisheries	 Centre.	 This	 model	represented	the	continental	shelf	ecosystem	of	southern	BC,	covering	an	area	of	approximately	30,000	 km2	 (Southern	BC	 shelf	model	 1996).	 	 Based	on	 the	 southern	BC	 shelf	model,	 Beattie	(1999)	built	two	preliminary	Hecate	Strait	(HS)	Ecopath	models	representing	the	ecosystems	of	the	early	1900s	and	early	1990s,	with	an	aim	to	“restore”	towards	the	past	ecosystem	(this	 is	referred	 to	 as	 the	 Back	 to	 the	 Future	 (BTF)	 approach	 (Pitcher	 2001)).	 The	 most	 notable	structural	changes	in	the	Hecate	Strait	models	compared	to	the	southern	BC	shelf	model	were:	(1)	 the	 study	 area	 was	 shifted	 north	 to	 include	 HS	 and	 Dixon	 Entrance,	 with	 a	 total	 area	 of	40,000	 km2;	 (2)	many	new	 functional	 groups	 such	 as	 lingcod	 (Ophiodon	elongatus),	 turbot	or	arrowtooth	 flounder	 (Atheresthes	 stomias),	 walleye	 pollock,	 flatfish	 (Pleuronectiformes),	juvenile	salmon	etc.	were	included	(yielding	a	total	of	25	functional	groups)	and	parameterized;	(3)	 several	 lower	 trophic	 level	 groups	 in	 the	 parent	 model	 were	 merged,	 e.g.	 amphipods,	euphausiids,	 chaetognaths,	 salps,	 and	 copepods	 were	 aggregated	 into	 a	 single	 zooplankton	A Revised EwE Model of Haida Gwaii 		 7	group;	 and	 (4)	Pacific	hake	was	deleted	 from	 the	HS	models,	 assuming	 that	 its	northernmost	distribution	 was	 south	 of	 the	 study	 area	 (Beattie	 1999).	 Furthermore,	 Beattie	 (2001)	constructed	 a	 late	 1990s	 HS	 ecosystem	 model	 based	 on	 the	 early	 1990s	 HS	 model	 for	 the	analysis	of	 “optimal	 size	and	placement”	of	marine	protected	areas	 (MPAs)	by	expanding	 the	size	 of	 the	 study	 area	 to	 nearly	 70,000	 km2	 (including	 Queen	 Charlotte	 Sound)	 and	 nearly	doubling	 the	 number	 of	 functional	 groups	 (giving	 a	 new	 total	 of	 49),	 of	 which	 many	 had	separate	juvenile	and	adult	“split-pool”	representation.	A	set	of	models	representing	different	time	periods	for	the	northern	BC	(NBC)	ecosystem	was	developed	based	on	 the	 late	 1990s	HS	model	 (Beattie	 2001)	 in	 an	 attempt	 to	 determine	 the	“optimum	 restorable	 biomass”	 for	 each	 functional	 group	 (the	 key	 goal	 of	 the	 BTF	 approach)	(Vasconcellos	&	Pitcher	2002).	The	periods	modelled	were	initially	1750,	1900	and	2000,	based	on	the	outcomes	of	a	BTF	science	workshop	held	at	St.	John’s	College,	UBC	in	September	2000.	The	model	reference	years	were	believed	to	represent	moments	of	 large-scale	change	in	 local	fisheries	 exploitation	 patterns	 from	 the	 pre-contact	 period	 to	 the	 present	 (Vasconcellos	 &	Pitcher	2002).	The	most	notable	revision,	 including	dynamic	 improvement,	of	these	NBC	models	(1750,	1900	and	2000)	were	 carried	out	by	Cameron	Ainsworth	as	part	of	his	PhD	work	 (Ainsworth	2006;	Ainsworth	et	al.	2002).	All	of	these	models	comprised	53	functional	groups,	including	11	juvenile	and	adult	“split-pools”	for	the	representation	of	trophic	ontogeny.	Ainsworth	also	built	an	NBC	model	 for	 the	year	1950	and	tuned/fitted	the	simulated	dynamics	of	 the	model	with	50-year-long	 time	 series	of	 catch,	 effort,	 biomass,	 and	 climate	 forcing	 (1950-2000).	 The	model	 tuning	process	 also	 extensively	 employed	 one	 of	 the	 least	 used	 Ecopath	 base	 parameters,	 namely	biomass	accumulation	(BA).	BA	is	a	production	term	that	allows	Ecopath	to	diverge	from	steady-state	assumptions	when	 its	 value	deviates	from	zero.	The	 vulnerability	 matrix	 was	scaled	 with	 the	 respective	 base	predation	 mortality	 of	 1750,	1900,	 and	 2000	 models	 and	transferred	 to	 the	 other	 models	assuming	 constancy	 in	 the	maximum	 predation	 mortality	that	a	prey	can	experience	from	a	particular	 predator.	Vulnerabilities	are	multipliers	that	determine	 the	 maximum	predation	 mortality	 a	 predator	can	 impose	on	 its	prey	compared	to	 the	 Ecopath	 base	 predation	mortality	 (Figure	 1.)	 (Christensen	et	al.	2008).	Figure	1.	Relationship	between	vulnerability	and	predation	mortality	in	Ecosim.	(Reproduced	 with	 permission	 from	 Villy	 Christensen	 from	 EwE	 help	 files	http://sources.ecopath.org/trac/Ecopath/wiki/EwEugVulnerabilitiesInEcosim)	2016  Fisheries Centre Research Reports 24(2) 		8	Adapting the NBC EwE model for the Haida Gwaii (HG) ecosystem The	 current	Haida	Gwaii	 (HG)	 ecosystem	model	 represents	 a	 further	 advance	 on	 the	 existing	NBC	model.	We	started	our	revision	on	the	framework	of	Ainsworth’s	NBC	2000	model,	mainly	because	the	NBC	model	was	built	on	the	basis	of	recent	scientific	data	in	contrast	to	the	other	historical	 models	 mentioned	 in	 the	 previous	 section,	 which	 were	 largely	 based	 on	 expert	opinion,	local	ecological	knowledge,	literature	reviews,	and	archaeological	evidence	(Ainsworth	2006;	Ainsworth	et	al.	2008a).	Moreover,	 the	NBC	2000	model	 carried	 the	 fitted	vulnerability	parameters	 for	 each	 predator-prey	 interaction	 based	 on	 the	 1950	model	 and	 the	 1950-2000	time	 series.	 Vulnerability	 is	 one	 of	 the	 most	 important	 parameters	 in	 the	 Ecosim	 dynamic	simulations,	since	it	establishes	the	degree	of	top-down	vs.	bottom-up	control	of	the	predator-prey	interaction	(Christensen	&	Walters	2004).	In	the	following	sections,	the	term	“NBC	model”	refers	 to	 the	Northern	 British	 Columbia	model	 for	 the	 year	 2000	 from	Ainsworth	 (2006)	 and	Ainsworth	et	al.	(2008a)	unless	stated	otherwise.	The	 revisions	 to	 the	 NBC	 model	 made	 in	 developing	 the	 present	 HG	 model	 can	 be	 broadly	categorized	into	5	stages:	(1)	EwE	version	upgrading;	(2)	delimitation	of	the	existing	modelling	(study)	 area;	 (3)	 parameterization	 of	 readjusted	 functional	 groups	 (in	 several	 stages);	 (4)	addition	of	new	fisheries/fleets;	and	(5)	incorporation	of	spatial	information	on	functional	group	distributions.	These	 revisions	 aim	 to	 produce	 an	 improved	 understanding	 of	 species	 distribution	 and	interactions	 as	 well	 as	 ecosystem	 structure	 and	 dynamics,	 and	 thus	 help	 to	 improve	 the	prevailing	management	practices	and	foster	sustainable	fisheries	around	Haida	Gwaii.	Methods   We	used	the	Ecopath	with	Ecosim	(EwE)	ecosystem	modelling	suite	to	build	our	HG	model.	With	its	three	modules—Ecopath,	Ecosim,	and	Ecospace—EwE	facilitates	the	construction	of	a	static	ecosystem	model	 (Ecopath)	 that	 can	 then	 be	 used	 to	 run	 time-based	 dynamic	 (Ecosim)	 and	spatial	 (Ecospace)	 simulations.	 EwE	has	 undergone	 a	massive	 capacity	 advancement	 over	 the	three	 decades	 (Christensen	 &	 Pauly	 1992;	 Christensen	 &	 Walters	 2004;	 Pauly	 et	 al.	 2000;	Walters	 et	 al.	 1997)	 since	 the	 pioneering	 work	 of	 Jeff	 Polovina	 at	 NOAA	 on	 a	 mass-balance	model	of	the	French	Frigate	Shoals	(Northwestern	Hawaiian	Islands)	food	web	(Polovina	1984).	The	detailed	workings	of	EwE	are	described	in	(Christensen	&	Walters	2004;	Christensen	et	al.	2008)	The	following	section	presents	the	key	aspects	of	the	modelling	routines.	Ecopath Modelling	 in	EwE	begins	with	creating	a	mass-balance	model	using	Ecopath	 to	obtain	a	 static	snapshot	 of	 the	 ecosystem	 under	 study.	 The	 underlying	 principle	 behind	 the	 “mass	 balance”	approach	is	to	balance	the	energy	flow	among	different	trophically	linked	functional	groups	by	solving	 a	 set	of	 simultaneous	 linear	 equations	 (one	equation	 for	 each	 functional	 group).	 That	requires	various	biological	parameters	such	as	biomass,	production	and	consumption	rates,	and	the	diet	composition	for	each	group	to	be	modelled.	Two	“master”	equations	allow	Ecopath	to	achieve	mass-balance	for	the	food	web.	The	first	Equation	(	1	)	ensures	energy	balance	among	the	groups	as:	A Revised EwE Model of Haida Gwaii 		 9	!! ∗ ! ! ! = !! + (!! ∗ ! ! ! ∗!"!")! + !! + !"! + !! ∗ ! ! ! ∗ !− !!! 	 (	1	)	where	subscripts	i	and	j	indicate	prey	and	predator	group,	respectively;	B	stands	for	biomass,	P	for	production,	Y	for	total	fishery	catch,	Q	for	consumption,	and	E	for	net	migration	rate;	DCji	is	the	fraction	of	prey	i	in	the	diet	of	predator	j;	BA	refers	to	biomass	accumulation;	and	EE	to	ecotrophic	efficiency	i.e.	the	fraction	of	group	mortality	explained	in	the	model.	The	second	“master”	Equation	(	2	)	explains	the	energy	balance	within	a	functional	group	as:	!"#$%&'()"# = !"#$%&'(#) + !"#$%&'(%)* + !"#$$%&%'#()* !""#	 (	2	)		Ecopath	efficiently	handles	(supports)	age-structured	modelling	using	the	“multi-stanza”	setup	to	represent	ontogenetic	shifts	 in	each	group’s	diet	and	 its	vulnerability	 to	each	predator	and	fishery.	The	biomass	dynamics	between	juvenile	and	adult	groups	are	governed	by	the	Deriso-Schnute	delay-difference	model	(Christensen	&	Walters	2004).	The	 following	equations	 from	Walters	et	al.	 (2010)	describe	 the	biomass	dynamics	 for	groups	with	multiple	stanzas:	 !!!!,!!! = !!,!exp (−!!,!12 )	 (	3	)		 !!!!,!!! = !!!!,! + !!!,!	 (	4	)		!!,! = !!,!!!,!!!!!(!)!!!!(!) 	 (	5	)		In	the	above	set	of	equations,	the	Nat	represents	the	numbers	at	age	a	(expressed	in	months)	at	time	t,	calculated	based	on	mortality	in	any	given	stanza	Zs.	Wat	represents	the	weight	at	age	a	and	time	t.	Full	details	of	the	calculation	of	Wat	can	be	found	in	(Walters	et	al.	2010).	Ecosim The	static	mass-balance	Ecopath	model	is	then	used	to	initiate	a	time-based	dynamic	simulation	in	Ecosim	to	track	changes	in	the	biomass	of	functional	groups	with	temporal	changes	in	catch	patterns,	 food	web	 structure	 (predator-prey	 interactions)	 and	 environmental	 conditions.	 The	dynamic	simulation	is	defined	using	this	equation,	derived	from	the	first	“master”	equation	of	Ecopath:	 !"!!" = !! !!"! − !!"! + !! − !"! + !! + !! ∗ !! 	 (	6	)		where	dBi/dt	is	the	biomass	growth	rate	of	group	i	in	time-interval	dt;	gi	denotes	net	growth	efficiency	(P/Q)	of	group	i;	Qji	is	the	consumption	by	group	i	while	Qij	is	that	of	group	i	by	predators;	Ii	refers	to	the	immigration	rate;	MOi	explains	“other	mortality”	(excluding	fishing	and	predation),	Fi	designates	the	fishing	mortality	and	ei	is	the	emigration	rate.	2016  Fisheries Centre Research Reports 24(2) 		10	Consumption	 by	 a	 functional	 group	 highly	 depends	 on	 the	 available	 biomass	 of	 prey	 and	 its	exchange	rate	between	unavailable	and	available	states,	as	described	by	foraging	arena	theory	(Walters	 et	 al.	 1997).	 This	 theory	 states	 that	 each	 prey	 population	 is	 split	 into	 fractions	‘vulnerable’	 and	 ‘invulnerable’	 to	 a	 given	 predator,	 and	 the	 exchange	 rate	 between	 the	 two	fractions	 defines	 the	 vulnerability	 parameter	 for	 that	 predator-prey	 interaction	 (Ahrens	 et	 al.	2012;	 Christensen	 &	 Walters	 2004;	 Walters	 et	 al.	 2000).	 Each	 predator-prey	 interaction	 is	assigned	 a	 vulnerability	 (v)	 value,	 from	one	 to	 infinity.	 If	 v	 =	 1,	 a	 bottom-up	 or	 donor-driven	relationship	 is	 implied.	 Assigning	 a	 high	 value	 of	 v	 implies	 a	 top-down	 or	 predator-driven	interaction,	 in	which	predation	mortality	 is	 proportional	 to	 the	product	 of	 prey	 and	predator	abundance	(i.e.	a	Lotka-Volterra	model).	 It	also	implies	a	high	flux	rate	for	prey	species	 in	and	out	of	vulnerable	biomass	pools.	When	a	very	high	value	is	set	for	the	vulnerability	parameter,	if	the	 predator	 biomass	 doubles,	 the	 predation	 mortality	 would	 increase	 by	 approximately	 a	factor	of	two.	If	v	=	1,	a	similar	increase	in	predator	biomass	will	not	have	a	large	effect	on	the	predation	mortality.	Ecospace Ecospace	is	the	dynamic	spatial	module	of	Ecopath	with	Ecosim	(Pauly	et	al.	2000;	Walters	et	al.	2010;	Walters	et	al.	1999).	Ecospace	creates	a	spatial	grid	on	which	the	species	distributions	are	mapped.	 Habitat	 suitability	 for	 different	 functional	 groups	 is	 mapped	 using	 habitat	 capacity	maps.	Within	each	cell	 in	 the	spatial	grid,	a	 foraging	arena	 is	described	 for	possible	 food	web	interactions	 depending	 on	 presence	 and	 absence	 of	 species.	 The	 carrying	 capacity	 of	 the	foraging	 arena	 relationships	 is	 related	 to	 habitat	 suitability	 described	 in	 the	 habitat	 capacity	maps.	Designing the Ecospace map The	distribution	of	each	functional	group	is	delineated	on	a	two-dimensional	map	of	the	model	area	and	the	biomass	of	the	functional	group	at	any	point	in	that	area	is	scaled	according	to	the	distribution	map.	In	EwE	5,	the	map	was	divided	into	different	habitat	types	and	each	functional	group	 was	 associated	 with	 one	 or	 more	 habitats.	 For	 example,	 nearshore	 species	 would	 be	associated	with	 shallow	coastal	waters	and	wide-ranging	pelagic	 species	would	be	allowed	 to	roam	 within	 the	 pelagic	 habitats	 specified	 in	 the	 map.	 Thus,	 the	 biomass	 distribution	 was	determined	from	presence/absence	across	a	habitat	map.	Therefore,	each	cell	 in	the	map	was	either	 completely	 suitable	 or	 entirely	 unsuitable	 habitat	 for	 a	 functional	 group,	 with	 no	possibility	of	representing	intermediate	states.		In	the	current	(EwE	6)	version	of	Ecospace	(Christensen	et	al.	2014),	the	definition	of	functional	group	 spatial	 distributions	 has	 been	 greatly	 enhanced.	 This	 version	 brings	 together	 facets	 of	species	distribution	modelling	and	 trophic	 interactions	 into	a	 single	dynamic	 framework.	Each	functional	group	has	a	“GIS-like”	 layer	referred	to	as	a	habitat	capacity	map	(Steenbeek	et	al.	2013).	 For	 each	 cell	 in	 the	map,	 habitat	 capacity	 ranges	 from	 0	 to	 1.	 A	 value	 of	 0	 indicates	entirely	 unsuitable	 habitat	 while	 a	 1	 indicates	 the	 best	 possible	 habitat	 for	 the	 species	 and	values	 in	 between	 indicate	 different	 levels	 of	 habitat	 suitability.	 The	 earlier	 problem	 of	 not	being	 able	 to	 represent	 partially	 suitable	 habitats	 was	 solved	 with	 this	 modification.	 For	example,	 a	 nearshore	 species	 could	 have	 high	 habitat	 capacity	 in	 coastal	 waters	 and	 the	capacity	could	decrease	gradually	as	the	depth	of	the	water	column	increases.	Furthermore,	the	model	 allows	 the	 habitat	 capacity	 to	 be	 calculated	 as	 a	 function	 of	 bathymetric	 and	A Revised EwE Model of Haida Gwaii 		 11	environmental	 variables	 (Christensen	et	al.	2014).	 For	example,	 if	 the	depth	and	 temperature	preferences	of	 a	 species	 are	 known	and	 the	depth	and	 temperature	profiles	of	 the	modelled	area	are	available,	this	information	can	be	combined	into	a	habitat	capacity	map	(Figure	2).	This	functionality	 can	 be	 used	 to	 improve	 habitat	 capacity	 definitions	 for	 species	 for	 which	 only	presence/absence	information	and	environmental	preferences	are	available.			The	 biomass	 of	 a	functional	 group	is	 scaled	 over	 the	map	 according	 to	the	 habitat	capacity	 gradient.	For	 example,	 in	the	 base	Ecospace	 map,	the	 biomass	density	 (t.km-2)	of	a	 species	 with	 a	narrow	distribution	 along	the	 coast	 would	be	much	higher	in	each	cell	than	the	base	 Ecopath	biomass	 density	due	 to	 the	concentration	 of	biomass	 in	 a	relatively	 small	area.	 Conversely,	for	a	species	with	a	wider	distribution,	the	cell	biomass	density	would	be	closer	to	the	Ecopath	base	level.	Furthermore,	there	are	options	to	simulate	the	directional	movement	of	juvenile	and	adult	functional	groups	in	Ecospace	using	features	such	as	advection	and	migration.	Movement between cells During	 an	 Ecospace	 run,	 an	 Ecosim	 simulation	 is	 enacted	 in	 each	 cell	 of	 the	 map.	 Biomass	distribution	 inside	 the	 unit	 cell	 (area	 =16	 km2	 in	 the	 present	 HG	 model)	 is	 considered	homogenous.	 At	 the	 end	 of	 the	 monthly	 time	 step,	 movement	 into	 the	 adjacent	 cells	 is	determined	by	 (1)	 the	base	dispersal	 rate	of	a	 functional	group,	 (2)	 the	 type	of	habitat	 in	 the	adjacent	 cell,	 (3)	 food	 availability	 and	 (4)	 predation	 risk.	 Equation	 (	 7	 )	 calculates	 the	 total	movement	out	Bout	(emigration)	from	a	cell.	Depth Temperature Habitat	CapacityFigure	2.	Habitat-based	foraging	capacity.	Map	of	 layers	 for	 depth	and	 temperature	and	an	 example	of	 habitat	 capacity	 calculated	based	on	species	preferences	for	depth	and	temperature	values.	2016  Fisheries Centre Research Reports 24(2) 		12	!!"#,!"# = !!,!!!"#!!!! 	 (	7	)		where	Bout	represents	the	biomass	density	of	group	i	exiting	the	cell	defined	by	row	r	and	column	c,	m	is	the	movement	rate	in	the	four	cardinal	directions	(N,	S,	E,	W)	represented	by	d.		Therefore,	the	net	biomass	flux	into	a	cell	 is	the	sum	of	the	fluxes	out	of	the	four	surrounding	cells.	 The	 base	 dispersal	 rate	 defines	 the	 annual	 net	 movement	 between	 cells	 for	 each	functional	 group,	 and	 this	 movement	 is	 adjusted	 according	 to	 the	 habitat	 capacity	 of	 the	adjacent	cell	such	that	there	is	a	negative	gradient	against	moving	into	cells	with	poorer	habitat	capacity.	 Thus,	 in	 Ecospace	movement	 rates	 are	 adjusted	 such	 that	 the	 abundance	 gradients	created	based	on	the	capacity	maps	are	maintained,	i.e.	movement	based	on	dispersal	rates	is	not	allowed	to	override	the	spatial	distribution	of	the	species	(Christensen	et	al.	2014).	Spatial representation of fishing effort All	 fisheries	 (fleets)	present	 in	 the	Ecopath	model	and	Ecosim	simulations	are	carried	 forward	into	 Ecospace.	 The	 spatial	 distribution	 of	 each	 fleet’s	 fishing	 effort	 on	 the	 Ecospace	 map	 is	governed	at	three	levels	(1)	fishing	fleets	are	associated	with	habitats	(for	example	prawn	traps	in	B.C.	operate	in	depths	up	to	100m);	(2)	fisheries	inside	the	map	can	be	spatially	restricted	by	setting	up	marine	protected	areas	(MPAs)	and	when	all	fisheries	are	excluded	from	an	MPA,	it	represents	a	no-take	area;	(3)	following	the	constraints	set	up	by	associating	fleets	with	habitats	and	defining	MPAs,	sailing	costs	can	be	used	to	improve	the	spatial	distribution	of	fishing	effort.	There	 are	 two	mechanisms	 within	 Ecospace	 to	 achieve	 this.	 Firstly,	 the	 user	 can	 specify	 the	location	of	ports	on	the	base	map,	and	Ecospace	will	 then	calculate	the	spatial	distribution	of	effort	assuming	an	ideal	free	distribution	based	on	distance	from	the	coast.	Effort	concentrates	in	 areas	 that	 have	 low	 sailing	 costs	 or	 are	 highly	 profitable	 owing	 to	 the	 high	 abundance	 of	target	species.	The	second	option	is	to	provide	directly	a	map	of	the	sailing	cost	for	the	region	based	on	information	gathered	from	fishers,	observers	or	surveys.		Setting up scenarios Spatial	management	options,	including	spatial	closures	such	as	marine	protected	areas	(MPAs),	can	be	set	up	using	Ecospace.	Forms	are	available	 to	set	up	the	number	of	MPAs.	The	spatial	extent	of	each	MPA	can	then	be	designated	on	the	base	map.	Once	the	MPAs	are	designated,	fisheries	 can	 be	 fully	 excluded	or	 partially	 restricted	 there.	 Partial	 restrictions	 are	 effected	 in	terms	of	the	number	of	months	fisheries	are	allowed	to	operate	inside	the	MPAs.	Furthermore,	the	effort	of	a	particular	fleet	can	be	scaled	down	within	Ecospace	simulations	using	functions	within	Ecosim.	The	spatial	biomass	dynamics	in	the	areas	surrounding	MPAs	can	be	examined	to	investigate	 the	 dynamics	 of	 biomass	 and	 catch	 in	 spillover	 areas.	 It	 is	 possible	 to	 obtain	 the	biomass	 dynamics	 results	 for	 any	 cell	 in	 the	 Ecospace	map,	 but	 often	 users	 are	 interested	 in	obtaining	average	results	for	MPAs	or	other	areas	of	interest	(e.g.	spillover	zones).	In	this	case,	users	can	establish	multiple	regions	in	Ecospace	for	which	average	results	can	be	obtained.	Several	 studies	 have	 used	 EwE	 5	 to	 explore	 spatial	 management	 options.	 Examples	 in	 B.C.	include	work	on	northern	BC	(Ainsworth	2006;	Ainsworth	et	al.	2008b)	and	more	specifically	on	Gwaii	Haanas	Marine	Conservation	Area	Reserve	(Salomon	et	al.	2002).	Other	examples	include	A Revised EwE Model of Haida Gwaii 		 13	Hong	Kong	(Pitcher	et	al.	2002),	the	Faroe	Islands	(Zeller	&	Reinert	2004),	the	northern	Adriatic	(Fouzai	et	al.	2012),	the	Central	Pacific	(Martell	et	al.	2005),	and	Raja	Ampat,	Indonesia	(Varkey	et	al.	2012).	Previous	studies	found	that	the	response	of	species	to	marine	protected	areas	was	dependent	on	multiple	factors	such	as	dispersal	rate,	habitat,	and	spillover	fisheries	(Varkey	et	al.	2012).	EwE version upgrading EwE	 has	 gone	 through	 several	 functionality	 improvements	 to	 address	 contemporary	management	challenges	since	its	advent	approximately	30	years	ago.	EwE	6	(Christensen	et	al.	2008)	 is	the	most	recent	release	of	the	modelling	suite,	and	we	have	utilized	a	number	of	the	advanced	capabilities	of	this	version	to	model	the	HG	marine	ecosystem.	The	NBC	model	adapted	 for	 the	 study	was	built	 in	EwE	5,	an	earlier,	obsolete	version	 lacking	several	 important	features.	For	example,	 it	had	a	limited	capability	to	model	trophic	ontogeny	using	the	juvenile	+	adult	“split-pool”	and	was	incapable	of	defining	group	spatial	distributions	using	habitat	capacity	maps.	Because	of	technical	difficulties,	especially	with	the	export	of	the	groups	having	juvenile	and	adult	representation,	the	NBC	model	was	manually	rebuilt	in	EwE	6	to	serve	as	the	basis	for	the	HG	model.	Delimitation of the study area The	present	study	area	retained	most	of	 the	original	geographical	extent	modelled	by	Beattie	(2001)	 and	 Ainsworth	 et	 al.	 (2002).	 However,	 in	 order	 to	 focus	 on	 waters	 around	 HG,	 the	southern	boundary	of	the	model	area	was	moved	north	(to	~70	km	south	of	Cape	St.	James,	the	southern	tip	of	HG),	and	the	western	boundary	was	extended	further	into	open	waters	(Figure	3).	 Thus,	 the	 total	 water	area	 modelled	 in	 the	study	 is	now	81,008	km2,	including	 Dixon	 Entrance	and	 a	 small	 part	 of	Southeast	 Alaska	 waters	in	 the	 north,	 Hecate	Strait	 and	 part	 of	 the	Inside	 Passage	 (Chatham	Sound)	 in	 the	 east,	 the	northernmost	 part	 of	Queen	 Charlotte	 Sound	in	 the	 south,	 and	 open	Pacific	 waters	 in	the	 west.	 This	study	 area	approximately	 corresponds	 to	 Pacific	 Fishery	 Management	 Areas	 101	 and	 103	 in	 the	 north,	Areas	102	and	104-107	in	the	east,	part	of	Area	130	in	the	south,	and	finally	Area	142	and	part	of	101	in	the	west	(DFO	2013a).		Figure	3.	Study	area.		The	figure	on	the	right	is	the	representation	in	Ecospace	of	the	actual	map	(left)	of	the	model	area	2016  Fisheries Centre Research Reports 24(2) 		14	Parameterization of readjusted functional groups  In	spite	of	the	aggregation	of	juveniles	and	adults	for	many	groups,	the	disaggregation	of	some	of	the	original	NBC	model	groups	(especially	for	marine	mammals	and	elasmobranchs)	and	the	addition	of	a	new	group	for	Pacific	hake	increased	the	total	number	of	functional	groups	in	the	model	from	the	53	found	in	the	NBC	model	to	56	(Figure	5).	Out	of	a	total	11	species	that	were	modelled	as	juvenile	and	adult	life	stages	in	the	NBC	model,	the	juvenile	and	adult	split	pools	of	9	species	(18	functional	groups)	were	combined	into	one	age	group	each,	while	the	remaining	two	species,	Pacific	halibut	(Hippoglossus	stenolepis)	and	Pacific	herring	(Clupea	pallasii),	were	modelled	using	the	multi-stanza	approach.	The	changes	made	to	the	groups	in	the	NBC	model	(merging	 split	 pools,	 adding	 new	 groups)	 necessitated	modifications	 to	 the	 diet	 composition	matrix.	For	the	merged	split-pools,	the	diets	were	rescaled	so	that	the	merged	groups	together	exerted	the	same	level	of	predation	pressure	on	their	prey	as	they	did	before	merging;	similarly,	the	merged	groups	together	contributed	the	same	amount	to	predator	diets	as	before	merging.	For	the	new	groups,	diets	were	created	based	on	published	data,	as	described	below	for	each	group.	 The	 altered	diet	 matrix	 is	presented	 in	 (Figure	8).	 The	 production	per	 biomass	 (P/B)	and	consumption	per	biomass	 (Q/B)	 values	of	 each	 merged	group	 were	 scaled	based	 on	 the	biomasses	 of	juveniles	 and	 adults	in	 the	 original	representation.	Figure	 4	 shows	 the	biomass	 spectrum	 of	the	HG	model’s	 food	web	 while	 Figure	 5	details	 the	 origin	 of	each	 of	 the	 model’s	functional	group.		Figure	4.	Biomass	spectrum	of	the	Haida	Gwaii	food	web	A Revised EwE Model of Haida Gwaii 		 15	2016  Fisheries Centre Research Reports 24(2) 		16	A Revised EwE Model of Haida Gwaii 		 17		Figure	5.	A	comparison	of	NBC	ecosystem	model	and	HG	ecosystem	model	functional	groups	and	Ecopath	parameters	Functional	groups	Ecosystem	players	starting	from	the	lowest	trophic	levels	of	the	food	web	such	as	producers	to	the	 highest	 trophic	level	 predators	 such	as	marine	mammals,	were	represented	by	the	 56	 functional	groups	 in	 the	 HG	model	 (Figure	 6).	 In	the	 following	section,	 we	 briefly	described	 the	parameterization	 of	each	 of	 the	functional	 groups.	We	also	recommend	interested	 readers	to	 look	at	Ainsworth	(2006)	 for	 an	 in-depth	 description	 of	the	 inherited	functional	 groups	from	 the	 NBC	model.	Figure	6.	Plot:	functional	groups	against	trophic	levels.	All	 56	 functional	 groups	 of	 the	 HG	 ecosystem	model	 plotted	 categorically	 against	 their	 trophic	levels:	 each	node	represents	 a	 functional	 group,	and	the	size	of	each	node	 is	 proportional	 to	 its	biomass	at	log-scale.	2016  Fisheries Centre Research Reports 24(2) 		18	Sea	otters	(Enhydra	lutris)	The	 only	 fully	 marine	 mustelid,	 the	 sea	 otter	 is	 a	 keystone	 in	 the	 kelp	 forest	 ecosystem,	exercising	top-down	control	over	herbivores	such	as	sea	urchins	(Estes	&	Palmisano	1974).	The	lucrative	maritime	fur	trade	in	the	18th	and	19th	centuries	reduced	the	global	abundance	of	this	species	 to	 several	 thousand	 by	 the	 early	 1900s,	 when	 hunting	 was	 terminated	 by	 an	international	 treaty,	 by	 which	 time	 uncontrolled	 grazing	 had	 had	 severe	 impacts	 on	 North	Pacific	kelp	forests.	Due	to	the	combined	efforts	of	various	agencies	towards	the	re-introduction	and	 protection	 of	 sea	 otters,	 the	 populations	 in	 many	 parts	 of	 the	 North	 Pacific	 started	 to	rebound	in	the	late	20th	century,	with	an	average	rate	of	increase	of	17-20%	per	year	in	parts	of	British	Columbia	(Blood	1993).	This	species	has	yet	to	permanently	recolonize	the	model	area	in	appreciable	 numbers,	 although	 vagrant	 individuals	 have	 been	 sighted	 (Raum-Suryan	 et	 al.	2004).	In	the	HG	model,	we	kept	this	group’s	Ecopath	parameters	(B,	P/B,	Q/B,	and	diet)	from	the	NBC	model	unchanged	(Ainsworth	2006).	For	modifications	to	the	representation	of	the	sea	otter	–	sea	urchin	–	kelp	trophic	cascade	in	the	model	to	be	implemented	in	the	next	revision	of	the	HG	model,	please	see	Part	B	below.	Baleen	whales	(Mysticeti)	The	 single	 group	 “Mysticetae”	 encompassing	 all	 such	 species	 in	 the	 original	 NBC	 model	(Ainsworth	et	al.	2008a)	was	disaggregated	to	species	level	in	the	HG	model	for	the	purposes	of	the	study	described	in	(Surma	&	Pitcher	2015)	and	for	the	sake	of	greater	ecological	realism.	Gray	whales	(Eschrichtius	robustus)	This	species,	the	only	bottom-feeding	baleen	whale,	passes	through	the	model	area	en	route	to	feeding	 grounds	 in	 Alaska.	 It	 is	 believed	 that	 a	 small	 “resident”	 population	 does	 not	migrate	further	north	and	feeds	in	British	Columbia	waters	throughout	the	summer	(Heise	et	al.	2003)		The	biomass	density	 (B)	 value	 for	 this	 group	was	obtained	 from	 the	 remainder	of	 the	baleen	whale	 biomass	 density	 in	 the	NBC	model	 (Ainsworth	 et	 al.	 2008a)	 after	 all	 other	 species	 had	been	accounted	 for.	This	value	was	 then	modified	upward	to	reflect	 the	moderately	common	occurrence	of	gray	whales	in	Haida	Gwaii	waters	(Heise	et	al.	2003).	The	P/B	and	Q/B	values	for	this	species	were	derived	from	(Guénette	2005).	The	source	P/B	values	were	raised	slightly	to	account	 for	 the	 very	 high	 rate	 of	 increase	 observed	 in	 the	 Northeast	 Pacific	 gray	 whale	population	 during	 its	 successful	 recovery	 from	 historical	 whaling	 (Reilly	 1984).	 The	 diet	composition	of	this	group	was	derived	from	the	original	“Mysticetae”	group	in	the	NBC	model	(Ainsworth	et	al.	2008a),	whose	biomass	was	dominated	by	this	species,	with	modifications	to	reflect	the	removal	of	other	baleen	whales	from	the	group.	Humpback	whales	(Megaptera	novaeangliae)	This	species	is	likely	the	most	abundant	cetacean	(in	terms	of	biomass)	in	the	model	area.	It	is	currently	 undergoing	 a	 successful	 recovery	 from	 historical	 depletion	 by	 20th-century	 whaling	(Surma	&	Pitcher	2015).	The	B	value	for	this	group	was	derived	from	the	Minimum	Number	Alive	(MNA)	calculated	by	Nichol	et	al.	(2010)	based	on	several	years	of	sightings	recorded	from	Haida	Gwaii	waters.	Given	that	 the	MNA	 included	 individuals	 that	only	used	 the	model	 area	 intermittently	 (Nichol	et	 al.	A Revised EwE Model of Haida Gwaii 		 19	2010),	half	of	this	value	was	taken	here	as	a	likely	estimate	of	the	average	local	abundance	of	humpback	whales.	This	method	is	similar	to	that	used	by	Guénette	(2005)	for	her	EwE	model	of	Southeast	 Alaska.	 The	 resulting	 current	 abundance	 estimate	was	 converted	 to	 biomass	 using	the	mean	 individual	mass	 for	 this	 species	 given	 by	 Trites	 and	 Pauly	 (1998).	 The	 P/B	 and	Q/B	values	 for	 this	 group	were	 taken	 from	 the	NW	Atlantic	 EwE	models	 published	by	Aráujo	 and	Bundy	(2011).	The	diet	composition	for	this	group	was	largely	based	on	(Pauly	et	al.	1998).	Much	additional	aid	was	provided	by	a	suggestion	by	Ford	et	al.	(2009)	that	stomach	contents	records	from	British	Columbia	whaling	stations	(which	would	 indicate	a	diet	of	~90%	euphausiids)	might	be	biased	towards	the	 latter	by	the	fact	that	most	whaling	catches	were	made	well	offshore.	 It	must	be	admitted	here	that	the	relative	importance	of	various	forage	fish	groups	in	the	modelled	diet	of	these	 whales	 is	 somewhat	 speculative,	 although	 herring	 are	 often	 mentioned	 as	 a	 major	component	 of	 humpback	 whale	 diet	 in	 the	 Northeast	 Pacific	 (Ford	 et	 al.	 2009;	 NMFS	 2011,	2014).	 In	addition,	the	humpback	whale	diet	composition	 in	this	model	closely	resembles	that	used	in	a	similar	model	of	Prince	William	Sound,	Alaska	(Okey	&	Pauly	1998).	The	proportion	of	the	 diet	 ascribed	 to	 “import”	 (i.e.	 feeding	 outside	 the	model	 area)	 was	 derived	 from	 known	levels	 of	 humpback	 whale	 fidelity	 to	 summer	 feeding	 grounds	 (Nichol	 et	 al.	 2010)	 and	 from	seasonal	whaling	catch	data	in	Gregr	et	al.	(2000).	For	a	further	discussion	of	humpback	whale	diet	 and	modifications	 to	 its	 representation	 in	 the	next	 revision	of	 the	HG	model,	 please	 see	Part	B	below.	Minke	whales	(Balaenoptera	acutorostrata)	The	smallest	of	the	baleen	whales	occurring	in	the	Northern	Hemisphere,	this	species	is	also	the	only	one	never	to	have	been	the	target	of	whaling	in	the	Northeast	Pacific.	The	biomass	density	(B)	value	for	this	group	originated	from	the	abundance	estimate	published	for	northern	British	Columbia	by	Williams	and	Thomas	(2007).	This	estimate	was	converted	to	biomass	using	the	mean	individual	mass	for	this	species	given	by	(Trites	&	Pauly	1998).	The	P/B	and	Q/B	values	for	minke	whales	were	obtained	from	the	NW	Atlantic	EwE	models	published	by	Aráujo	and	Bundy	(2011).		In	the	absence	of	detailed	data	on	local	diet,	the	diet	composition	for	this	group	was	based	on	Pauly	et	al.	(1998).	The	relative	importance	of	herring	and	other	forage	fish	 in	 the	 diet	 is	 therefore	 conjectural.	 For	 a	 further	 discussion	 of	 minke	 whale	 diet	 and	modifications	 to	 its	 representation	 in	 the	 next	 revision	 of	 the	 HG	 model,	 please	 see	 Part	 B	below.	Blue	whales	(B.	musculus)	The	largest	animal	in	the	world,	the	blue	whale	now	maintains	only	a	small	fraction	of	its	local	abundance	from	the	period	before	commercial	whaling	(Surma	&	Pitcher	2015).	 It	 is	a	dietary	specialist,	feeding	almost	exclusively	on	euphausiids	in	deep	offshore	waters.	The	biomass	density	(B)	value	for	this	group	was	based	on	the	number	of	recent	sightings	in	the	model	area	(Calambokidis	et	al.	2009)	and	the	status	of	the	North	Pacific	population	as	assessed	by	COSEWIC	 (2002).	Based	on	 these	sources,	 the	current	 local	abundance	of	blue	whales	was	estimated	to	be	<10	individuals.	This	figure	was	converted	to	biomass	using	the	mean	individual	mass	for	this	species	given	by	Trites	and	Pauly	(1998).	The	P/B	and	Q/B	values	for	blue	whales	were	 obtained	 from	 an	 EwE	model	 of	 Atlantic	 waters	 off	 northwest	 Africa	 (Morissette	 et	 al.	2016  Fisheries Centre Research Reports 24(2) 		20	2010),	as	no	such	models	of	temperate	northern	ecosystems	included	this	species	as	a	distinct	functional	group.	In	the	absence	of	detailed	local	data,	the	diet	composition	for	this	group	was	based	on	Pauly	et	al.	(1998).	The	proportion	of	the	diet	ascribed	to	“import”	(i.e.	feeding	outside	the	model	 area)	was	 derived	 from	 the	 likely	membership	 of	 locally	 sighted	 individuals	 in	 the	California	population	(Calambokidis	et	al.	2009)	and	from	seasonal	whaling	catch	data	in	Gregr	et	al.	(2000).	Fin	whales	(B.	physalus)	The	second	largest	animal,	this	species	was	once	the	most	abundant	cetacean	in	the	model	area	(in	 terms	 of	 biomass),	 but	 is	 now	 only	 beginning	 to	 recover	 from	 severe	 depletion	 by	 20th-century	whaling	(Surma	&	Pitcher	2015).	The	 biomass	 density	 (B)	 value	 for	 this	 group	 was	 based	 on	 the	 calculations	 made	 for	 all	 of	northern	 British	 Columbia	 by	 Williams	 and	 Thomas	 (2007),	 opportunistic	 data	 from	 the	 BC	Cetacean	Sightings	Network	(COSEWIC	2005)	and	expert	input	(Trites,	2013,	pers.	comm.).	The	estimated	 local	abundance	was	converted	 to	biomass	using	 the	mean	 individual	mass	 for	 this	species	given	by	Trites	and	Pauly	(1998).	The	P/B	and	Q/B	values	for	fin	whales	were	obtained	from	the	NW	Atlantic	EwE	models	published	by	Aráujo	and	Bundy	(2011).	The	diet	composition	for	this	group	was	derived	largely	from	whaling-era	stomach	contents	records	published	by	Flinn	et	 al.	 (2002).	However,	based	on	data	 from	other	 regions	of	 the	North	Pacific	 (Mizroch	et	 al.	2009)	 and	 a	 similar	 logic	 to	 that	 employed	 by	 Ford	 et	 al.	 (2009)	 for	 humpback	 whales,	 the	modelled	 proportion	 of	 fish	 in	 the	 fin	 whale	 diet	 was	 increased	 slightly	 relative	 to	 that	established	by	Flinn	et	al.	 (2002).	The	proportion	of	the	diet	ascribed	to	“import”	(i.e.	 feeding	outside	the	model	area)	was	derived	from	seasonal	whaling	catch	data	in	Gregr	et	al.	(2000).	Sei	whales	(B.	borealis)	This	elusive	deep-water	species	was	once	quite	abundant	 in	the	model	area	(Surma	&	Pitcher	2015),	but	is	now	extremely	rare	as	a	result	of	20th-century	whaling.	The	biomass	density	(B)	value	for	this	group	was	based	on	the	number	of	recent	sightings	in	the	model	area	(DFO	2012a)	and	the	status	of	the	North	Pacific	population	as	assessed	by	COSEWIC	(2003)	and	the	US	National	Marine	Fisheries	Service	(NMFS	2011).	Based	on	these	sources,	the	current	 local	 abundance	of	 sei	whales	was	estimated	 to	be	no	more	 than	 several	 individuals.	This	figure	was	converted	to	biomass	using	the	mean	individual	mass	for	this	species	given	by	Trites	 and	 Pauly	 (1998).	 The	 P/B	 and	Q/B	 values	 for	 sei	whales	were	 obtained	 from	 the	NW	Atlantic	EwE	models	published	by	Aráujo	and	Bundy	(2011).	The	diet	composition	for	this	group	was	derived	 from	whaling-era	 stomach	 contents	 records	published	by	 Flinn	et	 al.	 (2002).	 The	proportion	of	 the	diet	ascribed	 to	 “import”	 (i.e.	 feeding	outside	 the	model	area)	was	derived	from	seasonal	whaling	catch	data	in	Gregr	et	al.	(2000).	Toothed	whales	(Odontoceti)	The	 single	 group	 “Odontocetae”	 encompassing	 all	 such	 species	 in	 the	 original	 NBC	 model	(Ainsworth	et	al.	2008a)	was	disaggregated	to	species	level	in	the	HG	model	for	the	purposes	of	the	study	described	in	(Surma	&	Pitcher	2015)	and	for	the	sake	of	greater	ecological	realism.	A Revised EwE Model of Haida Gwaii 		 21	Sperm	whales	(Physeter	macrocephalus)	The	 largest	 toothed	whale,	 this	 species	 is	 now	 less	 abundant	 in	 the	model	 area	 than	 it	 was	historically	due	to	20th-century	whaling	in	the	North	Pacific	(Surma	&	Pitcher	2015).	Many	of	the	individuals	frequenting	the	area	are	mature	males,	which	are	larger	and	more	piscivorous	than	the	females.	The	biomass	density	(B)	value	for	this	group	was	based	on	an	estimate	by	Gregr	(2004)	and	on	expert	 input	 (Trites	 2013,	 Gisborne	 2013,	 pers.	 comm.).	 Based	 on	 these	 sources,	 the	 current	local	 abundance	 of	 sperm	 whales	 was	 estimated	 to	 be	 ~100	 individuals.	 This	 figure	 was	converted	to	biomass	using	the	mean	individual	mass	for	this	species	given	by	Trites	and	Pauly	(1998).	The	P/B	and	Q/B	values	for	this	group	were	obtained	from	the	EwE	model	of	Southeast	Alaska	 built	 by	 Guénette	 (2005).	 The	 diet	 composition	 for	 sperm	 whales	 was	 derived	 from	whaling-era	 stomach	 contents	 records	published	by	 Flinn	et	 al.	 (2002).	 The	proportion	of	 the	diet	 ascribed	 to	 “import”	 (i.e.	 feeding	 outside	 the	 model	 area)	 was	 derived	 from	 seasonal	whaling	catch	data	in	Gregr	et	al.	(2000).	Resident	and	transient	orcas	(Orcinus	orca)	Resident	orcas	specialize	in	feeding	on	Pacific	salmon	(Oncorhynchus)	with	particular	emphasis	on	Chinook	salmon	(O.	tschawytsha),	while	transients	preferentially	feed	on	marine	mammals,	particularly	 harbor	 seals	 (Phoca	 vitulina).	 The	 two	 ecotypes	 are	 reproductively	 isolated,	 and	hostile	 interactions	between	them	have	been	observed,	although	mutual	 indifference	 is	more	common.	For	these	reasons,	each	ecotype	has	been	allocated	its	own	functional	group	in	the	HG	model.	The	 biomass	 density	 (B)	 value	 for	 these	 groups	 originated	 from	 the	 abundance	 estimates	published	 for	 northern	 British	 Columbia	 by	Williams	 and	 Thomas	 (2007).	 These	 figures	 were	converted	to	biomass	using	the	mean	individual	mass	for	orcas	given	by	Trites	and	Pauly	(1998).	The	P/B	and	Q/B	values	for	these	groups	were	obtained	from	the	EwE	model	built	for	Southeast	Alaska	by	Guénette	(2005).	The	diet	compositions	for	both	ecotypes	were	based	on	Ford	et	al.	(1998)		and	Matkin	et	al.	(2007)		Small	odontocetes	This	 group	 contains	 species	 such	as	Pacific	white-sided	dolphin	 (Lagenorhynchus	obliquidens),	northern	right	whale	dolphin	(Lissodelphis	borealis),	harbour	porpoise	(Phocoena	phocoena)	and	Dall’s	 porpoise	 (Phocoenoides	 dalli).	 These	 species	 are	 important	 predators	 of	 small	 fish	 and	squid,	both	inshore	and	offshore,	and	are	in	turn	hunted	by	transient	orcas.	The	biomass	density	 (B)	 value	 for	 this	 group	originated	 from	 the	abundance	estimates	 for	 its	constituent	 species	 published	 for	 northern	 British	 Columbia	 by	Williams	 and	 Thomas	 (2007).	These	figures	were	converted	to	biomass	using	the	mean	individual	mass	for	each	species	given	by	 Trites	 and	 Pauly	 (1998)	 and	 summed	 to	 yield	 the	 total	 group	 biomass.	 The	 P/B	 and	 Q/B	values	 for	 small	 odontocetes	were	 obtained	 from	 the	NW	Atlantic	 EwE	models	 published	 by	Aráujo	and	Bundy	(2011).	The	diet	composition	for	this	group	was	derived	from	Gregr	(2004).	Pinnipeds	(seals	and	sea	lions)	This	group	from	the	original	NBC	model	(Ainsworth	et	al.	2008a),	also	used	in	Surma	and	Pitcher	(2015),	was	disaggregated	in	the	second	stage	of	HG	model	adaptation	by	placing	seals	(mainly	2016  Fisheries Centre Research Reports 24(2) 		22	harbour	seals	but	also	northern	fur	seals,	Callorhinus	ursinus)	and	sea	 lions	(mainly	Steller	sea	lions,	 Eumetopias	 jubatus	 but	 also	 California	 sea	 lions,	 Zalophus	 californianus)	 in	 separate	groups	of	equal	biomass,	both	of	which	inherited	the	P/B	and	Q/B	values	of	the	parent	group.	The	biomasses	and	diet	compositions	of	these	groups	are	derived	from	Gregr	(2004).	Seabirds	This	group	represents	over	a	dozen	major	piscivorous	and	planktivorous	bird	species	in	northern	BC,	 such	 as	 Cassin’s	 Auklets	 (Ptychoramphus	 aleuticus),	 Rhinoceros	 Auklets	 (Cerorhinca	monocerata),	Marbled	Murrelets	 (Brachyramphus	marmoratus),	 other	 auklets	 and	murrelets,	Pigeon	 Guillemots	 (Cepphus	 columba),	 murres	 (Uria	 spp),	 cormorants	 (Phalacrocorax	 spp.),	loons	 (Gavia	 spp.),	 gulls	 (Larus	 spp.)	 and	 kittiwakes	 (Rissa	 spp.),	 etc.	 It	 was	 estimated	 that	Cassin’s	Auklets	comprised	nearly	50%	of	the	total	>5.6	million	seabirds	nesting	on	the	BC	coast	(Tranquilla	et	al.	2007).	The	 parameters	 of	 this	 functional	 group	 were	 kept	 unchanged	 from	 the	 2000	 NBC	 model	(Ainsworth	2006).	As	cited	in	Ainsworth	(2006),	the	biomass	density	value	for	this	group	(0.0074	t/km-2)	 was	 taken	 from	 Kaiser	 (2002),	 while	 P/B	 (0.1	 yr-1)	 and	 Q/B	 (105.2	 yr-1)	 values	 were	obtained	from	Wada	and	Kelson	(1996).	Forage	fish,	copepods,	and	euphausiids	are	the	major	prey	items	(together	>50%)	for	this	group.	Pacific	Salmon	(Oncorhynchus	spp.)	The	 five	species	of	Pacific	 salmon	 found	on	 the	BC	coast	are	 represented	 in	 the	HG	model	as	three	 functional	 groups,	 as	 they	were	 in	 the	NBC	model:	 transient	 salmon,	 coho	 salmon,	and	Chinook	 salmon.	 These	 anadromous	 species	 cross	 the	 study	 area	 during	migrations	 between	spawning	 ground	 in	 rivers	 and	 offshore	 feeding	 grounds	 in	 the	 Gulf	 of	 Alaska.	 Commercial	landings	 of	 Pacific	 salmon	 off	 the	 BC	 coast	 have	 declined	 tremendously	 in	 the	 1990s,	 from	nearly	 96,000	 tonnes	 in	 1990	 to	 merely	 17,000	 tonnes	 in	 1999.	 	 However,	 since	 then	 the	landings	 have	 been	 almost	 stable	 most	 of	 the	 year	 at	 around	 20,000	 tonnes	 (DFO	 2014).	Gillnetters	and	seiners	capture	a	 large	proportion	of	the	total	salmon	catches	on	the	BC	coast	(as	in	the	NBC	model),	although	trollers	also	target	these	species.	Transient	salmon	This	 group	 includes	 pink	 salmon	 (O.	 gorbuscha),	 chum	 salmon	 (O.	 keta),	 and	 the	 most	commercially	 important	sockeye	salmon	(O.	nerka).	 Its	parameters	were	kept	unchanged	from	the	NBC	model.		Ainsworth	(2006)	estimated	the	biomass	in	the	study	area	based	on	catch	data	as	0.208	t/km2.	The	 P/B	 (2.48	 yr-1)	 and	 Q/B	 (8.33	 yr-1)	 values	 were	 obtained	 from	 the	 literature:	 Newlands	(1998)	and	(Christensen	1996)	respectively.	As	they	are	migratory	species,	most	of	 the	diet	of	these	 salmon	 come	 from	 outside	 the	 study	 area	 (mainly	 the	 northern	 Gulf	 of	 Alaska),	 and	therefore	 Ainsworth	 (2006)	 assigned	 60%	 of	 this	 group’s	 diet	 to	 “import.“	 Catch	 data	 were	obtained	 from	 official	 catch	 data	 for	 salmon	 recorded	 by	 DFO,	 and	 30%	 of	 the	 total	 was	allocated	to	the	group	(Ainsworth	2006).		Coho	salmon	(O.	kisutch)	The	Ecopath	parameters	(B,	P/B,	and	Q/B)	for	coho	salmon	in	the	HG	model	are	the	same	as	in	Ainsworth	 (2006)	 and	 Beattie	 (2001).	 Their	 values	 are	 0.024	 t/km2,	 2.76	 yr-1,	 and	 13.8	 yr-1	A Revised EwE Model of Haida Gwaii 		 23	respectively.	Q/B	was	estimated	by	Ecopath	using	the	mass-balance	approach	with	a	given	P/Q	=	0.2.	Nearly	 80%	of	 the	 group	diet	 is	 composed	of	 invertebrates,	 especially	 euphausiids	 and	squids	 while	 forage	 fish	 provide	 the	 remainder.	 As	 for	 transient	 salmon,	 30%	 of	 commercial	landings	reported	by	DFO	under	the	“salmon”	category	were	assigned	to	the	group	(Ainsworth	2006).	Chinook	salmon	(O.	tshawytscha)	The	Fraser	River	system	is	one	of	the	few	watersheds	in	BC	where	Chinook	salmon	spawn.	This	species	 does	 not	migrate	 very	 far	 into	 the	open	ocean,	 staying	mostly	 in	 coastal	waters,	 and	therefore	it	becomes	the	most	vital	prey	for	resident	killer	whales	(DFO	2010a).		The	model	parameters	 for	 this	group	are	derived	directly	 from	Ainsworth	 (2006):	 the	B	value	(0.034	t/km-2)	was	based	on	“catch	and	escapement	data,”	the	P/B	(2.16	yr-1)	was	carried	over	from	Beattie	(2001)	and	Q/B	(13.8	yr-1)	was	estimated	by	Ecopath	based	on	a	given	P/Q	=	0.2,	as	for	coho	salmon.	Catch	data	were	obtained	from	official	catch	data	for	salmon	recorded	by	DFO,	and	40%	of	the	total	was	allocated	to	the	group	(Ainsworth	2006).	Euphausiids	and	forage	fish	such	as	capelin	(Mallotus	villosus)	and	sandlance	(Ammodytes	hexapterus)	comprise	most	of	the	diet	of	Chinook	salmon.	Euphausiids	are	especially	dominant	in	the	diet	of	juveniles	(Davis	et	al.	2009).	Squid	The	squid	found	off	the	BC	coast	were	placed	into	two	functional	groups:	small	squid	and	large	squid	 (Ainsworth	2006).	Small	 squid	 includes	opal	 squid	 (Loligo	opalescens),	which	 is	 found	 in	inshore	 waters	 off	 the	 BC	 coast	 (DFO	 1999c).	 Large	 squid	 comprise	 the	 oceanic,	 migratory	species	 that	 come	 to	 BC	 waters	 to	 feed	 during	 summer	 from	 their	 sub-tropical	 spawning	grounds	 (DFO	 1999b).	 This	 group	 includes	 neon	 flying	 squid	 (Ommastrephes	 bartramii),	schoolmaster	gonate	squid	or	 red	squid	 (Berryteuthis	magister),	nail	 squid	or	boreal	clubhook	squid	(Onychoteuthis	borealijaponica),	and	eight-armed	squid	(Gonatopsis	borealis).	The	Ecopath	parameters,	i.e.	B	(small:	1.09	t/km-2,	large:	0.765	t/km-2),	P/B	(both	6.02	yr-1),	and	Q/B	(both	34.67	yr-1),	as	well	as	the	diet	compositions	for	these	groups	were	obtained	from	the	NBC	model	(Ainsworth	2006).	Ratfish	(Hydrolagus	colliei)	and	Dogfish	(Squalus	suckleyi)	According	 to	 the	 transboundary	 trawl	 survey	 conducted	 in	 2001	 in	 the	 southern	 Strait	 of	Georgia,	ratfish	and	dogfish	were	the	two	most	abundant	bottom	fish.	The	former	constitutes	nearly	60%	and	the	latter	nearly	11%	of	the	total	bottom	fish	population	of	the	area	surveyed	(Palsson	et	al.	2003).	Dogfish	are	a	major	predator	of	squid	and	pelagic	fish	in	BC	waters.	All	 of	 the	Ecopath	parameters	 (B,	P/B	and	Q/B)	and	diet	 compositions	 for	 these	groups	were	inherited	from	Ainsworth	(2006)	and	Beattie	(2001).	Walleye	pollock	(Theragra	chalcogramma)	Walleye	 pollock	 is	 a	 cold-water,	 semi-pelagic	 fish	 distributed	 throughout	 the	 northeastern	Pacific	Ocean,	with	the	highest	abundance	occurring	 in	the	Bering	Sea.	The	Bering	Sea	pollock	fishery	is	also	considered	one	of	the	largest	fisheries	of	the	world	(Springer	1992).	The	data	from	a	 fishery-independent	 survey	 off	 the	 west	 coast	 of	 Vancouver	 Island	 reveal	 a	 significantly	2016  Fisheries Centre Research Reports 24(2) 		24	increasing	 biomass	 trend	 for	 pollock	 since	 2010	 (Chandler	 et	 al.	 2015).	 Pollock	 provide	 an	abundant	but	low-energy	prey	resource	for	many	predators,	including	marine	mammals	(Trites	2012,	pers.	comm.).	In	the	NBC	ecosystem	model,	the	biomass	of	this	species	was	split	into	two	groups:	juvenile	and	adult	pollock.	 In	 the	HG	model,	both	 juvenile	and	adult	 fish	were	merged	 into	a	 single	group	called	 “Pollock,”	 and	 therefore	 the	 biomass	 of	 the	 merged	 group	 is	 the	 simple	 sum	 of	 the	biomasses	of	the	groups	in	the	NBC	model.	The	combined	production	and	consumption	of	both	the	juvenile	and	adult	groups	were	divided	by	the	biomass	of	the	merged	group	to	obtain	the	P/B	and	Q/B	parameters	of	the	latter;	these	parameters	were	slightly	tuned	when	balancing	the	HG	model.	The	diet	composition	of	the	merged	group	was	calculated	by	weighting	the	diets	of	the	 component	 groups	 according	 to	 their	 total	 consumption	of	 each	prey	 item.	 	 The	diets	 of	predators	 that	 prey	 on	 pollock	 were	 also	 adjusted,	 typically	 by	 summing	 up	 the	 percentage	contributions	of	the	juvenile	and	adult	groups.	Forage	fish	and	Eulachon	(Thaleichthys	pacificus)	The	forage	fish	group	in	the	NBC	and	HG	models	is	composed	mostly	of	sandlance,	with	smaller	quantities	of	Pacific	 sardine,	northern	anchovy	 (Engraulis	mordax),	 capelin	and	various	 smelts	(Ainsworth	2006).	Because	of	its	cultural	importance	to	coastal	First	Nations,	Ainsworth	(2006)	separated	eulachon	from	the	“forage	fish”	of	Beattie’s	Hecate	Strait	model	(Beattie	2001)	and	assigned	1/6th	of	the	total	predation	pressure	of	the	forage	group	to	the	eulachon.	Eulachon	is	a	small	anadromous	forage	fish	distributed	only	along	the	northeast	Pacific	coast	from	northern	California	 to	 southern	 Alaska	 (Moody	 and	 Pitcher	 2010).	 Even	 though	 eulachon	 is	 not	 a	commercially	 important	 fish,	 they	 are	 socio-ecologically	 important.	 Ecologically,	 they	 provide	energy-rich	 food	 to	 a	 wide	 range	 of	 predators	 from	 fish	 to	marine	mammals	 and	 birds,	 and	socially	they	provide	staple	food	and	oil	 (also	called	“grease,”	an	 important	trade	 item	among	First	 Nations)	 to	many	 First	 Nations	 along	 the	 coast	 (Cambria	 Gordon	 Ltd.	 2006;	Moody	 and	Pitcher	2010).	COSEWIC	has	identified	three	populations	of	eulachon	in	BC:	Central	Pacific	coast,	Fraser	 River,	 and	 Nass/Skeena.	 All	 the	 three	 populations	 were	 assessed	 to	 have	 declining	abundances	(Chandler	et	al.	2015).	These	 two	groups	were	not	modified	 from	their	 representation	 in	Ainsworth	 (2006).	For	both	groups	 (forage	 fish	 and	 eulachon),	 biomass	 (B)	 values	 were	 estimated	 by	 Ecopath	 using	 an	ecotrophic	efficiency	(EE)	of	0.95,	while	P/B	and	Q/B	were	estimated	by	averaging	the	P/B	and	Q/B	 for	 juvenile	and	adult	herring,	 respectively.	 	Most	of	 the	diet	 for	both	groups	 consists	of	copepods	and	euphausiids,	along	with	smaller	quantities	of	jellyfish	etc.	(Ainsworth	2006).	Pacific	Ocean	Perch	(Sebastes	alutus)	Pacific	Ocean	perch	(POP),	a	semi-pelagic	fish,	distributed	from	southern	California	to	the	Bering	Sea,	 is	 the	 foremost	 rockfish	 species	 in	 the	 groundfish	 trawl	 fishery	 of	 the	 BC	 coast.	 With	average	annual	landings	of	around	5000	tonnes	coast	wide,	this	species	alone	constitutes	25%	of	 the	 total	 rockfish	 landings	 obtained	 by	 bottom	 trawling	 (DFO	 2013b).	 The	 lifespan	 of	 this	rockfish	can	be	as	long	as	a	century,	but	they	can	recruit	to	the	fishery	as	early	as	seven	years	old	(DFO	1999d,	2013b).	In	the	NBC	model,	POP	had	the	juvenile	and	adult	split-pool	representation,	and	the	biomasses	of	 these	groups	were	estimated	using	a	 catch-at-age	model	based	on	data	collected	near	 the	A Revised EwE Model of Haida Gwaii 		 25	Goose	 Island	 gully	 (Ainsworth	 2006),	 one	 of	 the	 major	 “historical”	 fishing	 areas	 in	 Queen	Charlotte	 Sound	 (DFO	 1999d).	 The	 invulnerable	 biomass	 from	 the	 catch-at-age	 model	 was	assigned	to	the	juvenile	group	whereas	the	vulnerable	biomass	was	assigned	to	the	adult	POP	group	(Ainsworth	2006).	Ainsworth	(2006)	used	the	same	P/B	and	Q/B	values	for	these	groups	as	in	Beattie’s	(2001)	Hecate	Strait	model.	In	the	HG	model,	the	 juvenile	and	adult	groups	were	merged,	and	all	 the	Ecopath	parameters	were	calibrated	in	a	similar	fashion	as	explained	above	for	walleye	pollock.		Other	rockfish	There	are	over	60	species	of	 rockfish	 (Sebastidae)	 found	along	 the	northeastern	Pacific	coast,	and	over	35	of	them	reside	in	BC	waters	(COSEWIC	2009).	Apart	from	the	Pacific	Ocean	Perch,	all	other	rockfish	(Sebastes	spp.)	were	placed	under	three	distinct	functional	groups	categorized	mainly	 based	 on	 their	 habitat	 and	 feeding	 habits:	 “Inshore	 rockfish,”	 “Piscivorous	 rockfish”	(juvenile	and	adult)	and	“Planktivorous	rockfish”	(juvenile	and	adult)	in	(Ainsworth	2006).		Inshore	rockfish		Inshore	rockfish	are	the	Sebastes	spp.	captured	mainly	by	hook	and	line	(DFO	2000b);	these	are	copper	 (S.	 caurinus),	 quillback	 (S.	 maliger),	 tiger	 (S.	 nigrocinctus),	 china	 (S.	 nebulosus)	 and	yelloweye	 rockfish	 (S.	 ruberrimus)	 (Ainsworth	 2006).	 The	 Ecopath	 parameters	 of	 the	 group	remained	unchanged	from	Beattie’s	(2001)	Hecate	Strait	model,	as	cited	in	Ainsworth	(2006).	Piscivorous	rockfish	This	 rockfish	 group	 in	 the	model	 represents	 those	 Sebastes	 spp.	 and	 Sebastolobus	 spp.	 that	largely	feed	on	fish	and	large	invertebrates,	such	as	rougheye	(Sebastes	aleutianus),	shortraker	(S.	borealis),	black	(S.	melanops),	blue	(S.	mystinus),	chillipepper	(S.	goodei)	and	dusky	rockfish	(S.	 ciliatus),	 shortspine	 thornyhead	 (Sebastolobus	 alascanus)	 and	 longspine	 thornyhead	 (S.	altivelis)	(Ainsworth	2006).		The	piscivorous	rockfish	were	split	into	juvenile	and	adult	groups	in	the	NBC	model,	but	in	the	present	 HG	 model,	 they	 were	 merged	 into	 a	 single	 group.	 The	 Ecopath	 parameters	 of	 the	merged	group	were	calibrated	using	a	similar	approach	as	was	used	for	walleye	pollock.		Planktivorous	rockfish	A	 number	 of	 plankton-feeding	 rockfish	 such	 as	 yellowmouth	 (Sebastes	 reedi),	 redstripe	 (S.	proriger),	 widow	 (S.	 entomelas),	 yellowtail	 (S.	 flavidus),	 darkblotch	 (S.	 crameri),	 canary	 (S.	pinniger),	splitnose	(S.	diploproa),	sharpchin	(S.	zacentrus),	Puget	Sound	(S.	emphaeus),	bocaccio	(S.	paucispinis)	and	shortbelly	rockfish	(S.	jordani)	were	included	in	this	group	(Ainsworth	2006).	This	group	was	also	split	 into	 juvenile	and	adult	biomass	pools	 in	 the	NBC	model	and	merged	into	a	single	group	in	the	present	HG	model.	The	Ecopath	parameters	of	the	merged	group	were	adjusted	as	described	for	walleye	pollock.		Turbot/Arrowtooth	flounder	(Atheresthes	stomias)	Turbot,	 also	 called	 arrowtooth	 flounder	 to	 distinguish	 it	 from	 the	Atlantic	 turbot,	 is	 a	 flatfish	found	from	Baja	California	to	the	Bering	Sea,	with	higher	abundance	in	its	northern	range	(Fargo	&	Starr	2001).	Most	of	the	fish	are	caught	in	the	groundfish	trawl	fishery	and	to	some	extent	in	the	groundfish	hook	and	line	fishery;	however,	a	large	proportion	of	the	harvest	is	discarded	at	2016  Fisheries Centre Research Reports 24(2) 		26	sea,	and	therefore	the	flatfish	fishery	is	economically	less	important	(Fargo	&	Starr	2001).	Even	though	 turbot	 plays	 a	 significant	 role	 as	 a	 predator	 of	 small	 fish	 in	 the	 Haida	 Gwaii	 marine	ecosystem,	 very	 few	 studies	 have	 been	 conducted	on	 its	 population	 dynamics	 (Fargo	&	 Starr	2001).		In	the	NBC	ecosystem	model,	Ainsworth	(2006)	placed	 juvenile	and	adult	turbot	 into	separate	functional	groups,	and	most	of	their	Ecopath	parameters	were	carried	over	from	Beattie	(2001).	However,	these	juvenile	and	adult	pools	were	merged	into	a	single	group	in	the	HG	model,	and	parameters	were	scaled	accordingly	as	described	above	for	walleye	pollock.	Flatfish	Other	than	turbot	and	halibut,	all	the	flounders,	soles,	and	sanddabs	(Citharichthys	spp.)	were	collectively	 placed	 in	 the	 group	 “Flatfish”	 in	 the	 NBC	 model	 (Ainsworth	 2006).	 Dover	 sole	(Microstomus	 pacificus),	 English	 sole	 (Parophrys	 vetulus)	 and	 rock	 sole	 (Lepidopsetta	 spp.)	contribute	 nearly	 80%	of	 the	 total	 flatfish	 landings	 in	 the	 groundfish	 trawl	 fishery	 off	 the	 BC	coast.	Two	stocks	of	Dover	sole,	northern	and	southern,	have	been	recognized	off	the	BC	coast;	Hecate	Strait	is	mostly	dominated	by	the	northern	stock	while	the	southern	stock	ranges	south	from	Queen	Charlotte	Sound	to	the	west	coast	of	Vancouver	 Island	 (DFO	1998).	Hecate	Strait	also	has	one	of	the	largest	English	sole	stocks	off	the	BC	coast	(DFO	1999a).	For	rock	sole,	four	populations	from	two	species	of	Lepidopsetta	are	found	in	the	present	study	area	(DFO	1999d).		In	the	NBC	model,	the	flatfish	group	was	split	 into	juvenile	and	adult	pools.	 In	the	present	HG	model,	we	have	merged	these	juvenile	and	adult	pools	into	a	single	group,	and	therefore,	all	the	Ecopath	 parameters	 for	 these	 groups	 were	 merged	 and	 rescaled	 as	 well,	 using	 a	 similar	approach	as	that	discussed	above	for	walleye	pollock.	Pacific	halibut	(Hippoglossus	stenolepis)	The	 maximum	 reported	 weight	 for	 Pacific	 halibut,	 one	 of	 the	 largest	 flatfish	 along	 with	 the	Atlantic	halibut,	is	318	kg.	However,	most	of	the	individuals	in	commercial	catches	weigh	<100	kg,	and	longline	is	the	main	commercial	gear	(IPHC	1998).	As	a	transboundary	resource,	which	is	found	 in	 the	 shelf	 waters	 of	 Canada	 and	 the	 USA,	 Pacific	 halibut	 is	 jointly	 managed	 by	 the	governments	 of	 Canada	 and	 the	 United	 States	 of	 America	 through	 the	 International	 Pacific	Halibut	Commission.	Halibut	are	important	predators	of	demersal	fish	and	large	invertebrates,	and	the	largest	individuals,	known	to	fishermen	as	“whales,”	have	few	natural	enemies	due	to	their	size	(>	2	m	in	length).	Ainsworth	(2006)	split	Pacific	halibut	into	juvenile	and	adult	“pools”	in	the	NBC	model,	and	the	Ecopath	 parameters	 such	 as	 B	 (juveniles:	 0.296	 t/km2,	 adults:	 0.608	 t/km2),	 P/B	 (0.4	 yr-1	 for	adults)	and	Q/B	(1.10	yr-1)	were	kept	similar	to	those	in	the	Hecate	Strait	model	(Beattie	2001).	In	 addition,	 based	on	 the	 assumption	 that	 the	 juvenile	 “pool”	was	more	productive	 than	 the	adult,	 P/B	 and	Q/B	 values	 for	 juveniles	 were	 increased	 to	 0.6	 yr-1	 and	 1.46	 yr-1,	 respectively	(Ainsworth	2006).	The	original	juvenile	and	adult	“pools”	of	the	NBC	model	were	converted	into	juvenile	and	adult	“stanzas”	in	the	HG	model.	The	multi-stanza	interface	of	Ecopath	requires	the	biomass	and	Q/B	parameters	for	any	one	age	group	(the	“leading	stanza”)	and	P/B	(which	is	assumed	to	be	equal	to	the	total	population	mortality	Z	under	equilibrium	conditions)	for	all	the	age	groups;	in	other	A Revised EwE Model of Haida Gwaii 		 27	words,	Ecopath	 restricts	 the	 input	biomass	and	Q/B	 to	a	 single	 life-history	 stage	 in	 the	multi-stanza	 interface.	 Ecopath	 then	 estimates	 biomass	 and	 Q/B	 for	 the	 juvenile	 stanza	 assuming	certain	underlying	conditions,	 such	as	a	 “stable	age-size	distribution”	 in	 the	given	population,	and	that	the	species	body	growth	follows	a	von	Bertalanffy	growth	curve	(Christensen	&	Walters	2004).	To	match	the	juvenile	halibut	biomass	in	both	the	HG	and	NBC	models,	we	increased	the	adult	biomass	by	nearly	50%	in	the	HG	model	compared	to	the	value	in	the	NBC	model.	We	did	not	make	any	significant	changes	in	the	halibut	P/B	values.		Both	the	juvenile	and	adult	groups	are	fished	in	the	HG	model	in	a	similar	fashion	as	in	the	NBC	model.	 The	 fisheries	 data	 input	 for	 the	 Ecopath	 model	 is	 based	 on	 a	 combination	 of	 IPHC	records	(2003),	estimated	IUU	catch,	and	known	recreational	catch	(Ainsworth	2006).		Pacific	cod	(Gadus	macrocephalus)	Pacific	 cod	 is	distributed	 from	southern	California	 to	 the	northern	Bering	Sea	and	also	on	 the	shelf	of	 the	Aleutian	 Islands	 (Fredin	1985);	 they	are	 thus	 found	off	 the	entire	 coast	of	British	Columbia.	Most	of	the	commercial	catches	come	from	Hecate	Strait	and	Queen	Charlotte	Sound	and	 are	 predominantly	 obtained	 by	 the	 groundfish	 trawl	 fishery	 (DFO	2015a).	 A	 recent	 stock	assessment	model	suggests	that	the	biomass	of	Pacific	cod	off	the	BC	coast	has	maintained	an	increasing	trend	since	2001,	“despite	large	uncertainty”	in	the	estimates	(DFO	2015a).		Pacific	 cod	was	 represented	 using	 juvenile	 and	 adult	 split	 pools	 in	 the	NBC	model.	 The	 adult	biomass	 and	 catches	were	 obtained	 from	 a	 stock-assessment	model	 by	 Sinclair	 et	 al.	 (2001),	while	 the	 juvenile	biomass	was	accounted	 for	by	assuming	 its	~	35%	contribution	to	 the	total	Pacific	 cod	 population	 (Ainsworth	 2006).	 In	 addition,	 Q/B	 parameters	 for	 both	 pools	 were	carried	 over	 from	 Beattie	 (2001).	 All	 Ecopath	 parameters	 for	 juvenile	 and	 adult	 cod	 were	rescaled,	as	described	above	for	walleye	pollock,	to	form	a	single	merged	group	called	“Pacific	cod”	in	the	HG	model.		Sablefish	(Anoplopoma	fimbria)	Sablefish	 is	a	demersal	and	often	extremely	migratory	 species.	There	are	year-round	 fisheries	for	sablefish,	and	most	of	the	catches	are	allocated	to	sablefish	trap	and	groundfish	trawl	with	some	 bycatch	 in	 the	 halibut	 hook	 and	 line	 fishery	 (DFO	 2011a).	 Stock	 assessment	 based	 on	management	 strategy	 evaluation	 (MSE)	 indicates	 sablefish	 spawning	 biomass	 to	 be	 slightly	below	BMSY	(DFO	2011a).	Unlike	 in	 the	NBC	model,	we	 have	 combined	 the	 juvenile	 and	 adult	 age	 classes	 into	 a	 single	sablefish	group	in	the	HG	model.	The	parameters	for	the	merged	group	were	adjusted	using	a	similar	approach	as	was	used	for	walleye	pollock.		Lingcod	(Ophiodon	elongatus)	Lingcod	 is	 considered	 a	 non-migratory	 species	 and	 distributed	 solely	 along	 the	 northeastern	Pacific	coast	from	Baja	California	to	the	Shumagin	Islands,	south	of	the	Alaska	Peninsula	(Cass	et	al.	1990;	DFO	2012b).	Lingcod	supports	commercial	as	well	as	recreational	fisheries	off	the	BC	coast,	and	the	majority	of	the	catches	come	from	trawl	and	hook	and	line	fisheries	(DFO	2012b).	Stock	 status	 in	 all	 four	 stock	 assessment	 areas,	 including	Hecate	 Strait	 (5C	 and	 5D)	 and	west	coast	of	Vancouver	Island	(5E)	in	BC,	which	cover	most	of	the	present	study	area,	is	assessed	to	be	in	“healthy”	zone	(≥	0.8BMSY)	(DFO	2012b).	2016  Fisheries Centre Research Reports 24(2) 		28	Lingcod	 was	modelled	 under	 two	 age	 classes,	 juvenile	 and	 adult	 lingcod,	 in	 the	 NBC	model,	which	were	aggregated	into	a	single	lingcod	group	in	the	present	HG	model.	Parameterization	of	the	merged	group	followed	the	same	procedure	as	described	above	for	walleye	pollock.		Shallow	water	benthic	fish	As	 in	 the	 NBC	model	 (Ainsworth	 2006),	 this	 group	 comprises	 a	 number	 of	 fish	 taxa	 such	 as	“sculpins	(Cottidae),	blennies	(Bleniidae),	poachers	(Agonidae),	gobies	(Gobieiedae),	greenlings	(Hexagrammidae,	 except	 lingcod),	 eelpouts	 (Zoarcidae),	 northern	 clingfish	 (Gobiesox	maeandricus),	 red	 Irish	 lord	 (Hemilepidotus	 hemilepidotus),	 cabezon	 (Scorpaenichthys	marmoratus),	 snowy	snailfish	 (Liparis	pulchrettus),	 cutthroat	 trout	 (Oncorhynchus	clarki	clarki)	and	white	sturgeon	(Acipenser	transmontanus)”.	Ecopath	input	parameters	in	the	HG	model	are	the	same	as	those	in	the	NBC	model.		Pacific	Herring	(Clupea	pallasii)		Pacific	 herring,	 which	 supports	 commercial	 as	 well	 as	 aboriginal	 fisheries	 in	 BC,	 plays	 a	 vital	ecological	role	by	contributing	 large	proportions	of	the	diets	of	a	number	of	predators	(Figure	7),	 such	 as	 Pacific	 hake,	 seals	 and	 sea	 lions,	 inshore	 rockfish,	 lingcod,	 humpback	 whales,	dolphins	 and	 porpoises,	 and	 seabirds,	 as	 well	 as	 by	 competing	 with	 other	 planktivores	 (e.g.	forage	 fish,	walleye	pollock)	 in	 the	NBC	 food	web.	On	 the	other	hand,	 the	abundance	of	 this	species	is	highly	unstable;	despite	being	managed	strategically	by	DFO,	its	lack	of	recovery	is	a	serious	cause	of	concern	to	First	Nations,	commercial	fishermen,	conservationists	and	scientists	alike.		The	Pacific	herring	population	 in	BC	waters	 is	believed	to	be	composed	of	 five	major	and	two	minor	stocks.	The	major	stocks	include	(1)	Haida	Gwaii	(HG),	(2)	Prince	Rupert	District	(PRD),	(3)	Central	Coast	(CC),	(4)	Strait	of	Georgia	(SOG)	and	(5)	West	Coast	of	Vancouver	Island	(WCVI).	The	minor	stocks	are	(1)	HG	Area	2W	and	(2)	WCVI	Area	27	(DFO	2015d).	Nearly	18%	of	the	total	length	of	the	BC	coastline	was	identified	as	suitable	spawning	habitat	for	Pacific	herring	based	on	an	analysis	of	nearly	30,000	spawning	events	over	the	last	85	years	(Hay	&	McCarter	2013).	In	 the	NBC	model,	all	of	 the	stocks	occurring	 in	 the	model	area	were	 represented	by	a	 single	herring	group	with	juvenile	and	adult	split	pools	(Ainsworth	2006).	In	the	present	HG	model,	the	split	pools	were	modified	 into	 juvenile	and	adult	stanzas	 for	a	single	herring	population,	as	 in	the	case	of	adult	halibut.	However,	since	data	collection	and	assessment	of	these	stocks	are	conducted	independently	by	DFO,	it	is	reasonable	to	model	these	stocks	separately	in	the	HG	ecosystem	model	so	that	DFO	catch	and	effort	data	 for	 the	different	 stocks	 can	be	correctly	applied	 to	 the	different	 stocks.	Also,	the	results	will	be	more	useful	for	comparative	purposes.		A Revised EwE Model of Haida Gwaii 		 29		Figure	7.	Predator	consumption	of	Pacific	herring	in	the	HG	ecosystem	model.	The	present	model	area	covers	most	of	the	distribution	ranges	of	three	major	stocks	(HG,	PRD,	and	CC)	and	one	minor	stock	(HG	Area	2W).	Therefore,	we	have	planned	to	model	these	stocks	independently	in	the	next	revision	of	the	HG	model.	In	other	words,	we	will	be	setting	up	four	multi-stanza	 functional	 groups	 for	 herring	 in	 the	 final	 HG	 model,	 in	 contrast	 to	 the	 single	functional	 group	 in	 the	 NBC	 model.	 Herring	 is	 a	 migratory	 fish,	 which	 seasonally	 migrates	between	 offshore	 feeding	 grounds	 and	 nearshore	 spawning	 areas	 (Hay	 1985).	 In	 BC,	 juvenile	herring	typically	recruit	to	the	spawning	population	at	age	3(DFO	2015d).	They	spawn	along	the	shoreline,	and	juveniles	spend	their	first	summer	in	nearshore	bays	and	channels	before	moving	to	offshore	feeding	grounds.	The	different	life	stages	differ	in	their	diet	composition	as	well	as	in	their	importance	to	the	diets	of	various	predators.	To	model	ontogenetic	shifts	in	feeding	style	and	 habitation,	 each	 of	 these	 functional	 groups	will	 be	 setup	with	 three	 stanzas	 in	 the	 next	revision	of	the	HG	model:	1. Young	of	year	herring	(ages:	0-12	months),	which	spend	most	of	their	 life	 in	nearshore	waters	2. Immature	 herring	 (ages:	 12-36	 months),	 which	 mostly	 stay	 offshore	 and	 join	 the	spawning	stock	at	the	end	of	the	stage	2016  Fisheries Centre Research Reports 24(2) 		30	3. Mature	 herring	 (ages:	 36+),	 the	 leading	 stanza,	 which	 undergoes	 seasonal	 migration	between	offshore	and	nearshore	waters	and	supports	most	of	the	herring	fisheries	such	as	roe	herring,	food	and	bait,	and	spawn-on-kelp.		Parameterization	 of	 stanzas	 in	 Ecopath	 requires	 the	 biomass	 and	 Q/B	 values	 of	 a	 “leading	stanza”	and	P/B	values	for	all	 the	stanzas	of	a	 functional	group	(Christensen	&	Walters	2004).	Biomass	and	total	mortality	(Z)	values	for	the	leading	stanza	(spawning	stock)	will	be	taken	from	the	 herring	 assessment	 report	 prepared	 by	DFO	using	 ISCAM	on	 a	 Bayesian	 framework	 (DFO	2015d)	and	(Cleary	2016,	pers.	comm.).	Pacific	Hake	(Merluccius	productus)	Pacific	 hake,	 also	 called	 Pacific	 whiting,	 was	 recently	 restored	 to	 the	 HG	 model.	 This	 semi-pelagic,	transboundary	migratory	fish	enters	BC	coastal	waters	between	late	spring	and	summer	to	 feed	 on	 forage	 fish	 and	 zooplankton	 and	 leaves	 the	 coast	 in	 fall	 heading	 for	 spawning	grounds	 off	 southern	 California	 (Beamish	 &	 McFarlane	 1985;	 DFO	 2009a).	 It	 is	 the	 largest	contributor	to	BC’s	total	groundfish	 landings,	exceeding	the	total	catch	of	all	other	groundfish	(DFO	2014).	Pacific	hake	was	included	in	the	southern	BC	shelf	model	(Southern	BC	shelf	model	1996),	but	it	was	 removed	 from	 the	 subsequent	 Hecate	 Strait	 and	 NBC	 models	 (Ainsworth	 et	 al.	 2002;	Beattie	1999,	2001)	on	the	assumption	that	the	northern	boundary	of	its	distribution	lay	south	of	 their	 study	 area.	 However,	 most	 likely	 because	 of	 climate	 change	 or	 decadal	 shifts	 in	oceanographic	 conditions,	 the	 summer	 range	 of	 Pacific	 hake	 has	 shifted	 to	 encompass	Haida	Gwaii	waters,	 and	 the	 species	was	 recently	 found	 to	 spawn	 off	 the	west	 coast	 of	 Vancouver	Island	 (McFarlane	 et	 al.	 2000).	 Furthermore,	 hake	 fisheries	 have	 also	 shifted	 north	 from	 the	traditional	 fishing	 grounds	 off	 southern	 Vancouver	 Island,	 and	most	 of	 the	 catch	 is	 obtained	using	midwater	 trawling	at	100-500	m	depths	 (Helser	et	al.	2008).	Considering	 the	northward	shift	in	its	range	and	its	increasing	presence	in	BC	waters,	Pacific	hake	were	reintroduced	in	the	HG	model.		Since	a	biomass	estimate	for	the	hake	present	within	the	HG	region	was	not	available,	we	made	the	following	assumptions	for	calculation	of	the	biomass:	• Based	on	the	hake	distribution	map	obtained	from	(Stewart	&	Hamel	2010,	p.	87,	figure	2),	we	assumed	that	hake	distribution	ranged	from	latitudes	55°	N	to	35.5°	N.	• We	assumed	that	90%	of	Pacific	hake	are	found	within	this	zone.	• We	assumed	that	the	distribution	is	concentrated	roughly	around	45°	N	and	the	biomass	concentration	decreases	northwards	and	southwards.	• We	 assume	 that	 the	 north-to-south	 concentration	 is	 similar	 to	 a	 standard	 normal	distribution	with	a	mean	around	45°	N.	• Based	on	the	assumption	of	normal	distribution,	the	area	under	the	curve	was	calculated	for	the	model	area	and	expressed	as	a	 fraction	of	the	area	under	the	standard	normal	distribution.	We	assumed	that	the	same	fraction	of	the	total	Pacific	hake	biomass	would	be	present	in	the	HG	region.	• Since	most	of	the	Pacific	hake	reside	in	BC	for	only	6	months,	the	value	calculated	in	the	above	step	was	halved	to	obtain	the	final	biomass	density	estimate	(0.8	t.km-2).	A Revised EwE Model of Haida Gwaii 		 31		In	the	Haida	Gwaii	ecosystem,	hake	is	an	important	predator	of	herring	(Schweigert	et	al.	2010).	Elasmobranchs	In	 the	second	stage	of	 improvements	 to	 the	HG	model,	 the	original	 functional	group	“Skates”	from	the	2000	NBC	model	(Ainsworth	et	al.	2008a),	containing	all	elasmobranchs	except	Pacific	dogfish	 (Squalus	 suckleyi),	 was	 split	 into	 four	 new	 functional	 groups:	 salmon	 sharks	 (Lamna	ditropis),	 blue	 sharks	 (Prionace	 glauca),	 large	 demersal	 sharks	 and	 small	 demersal	elasmobranchs	(small	sharks	and	skates).	Small	demersal	elasmobranchs	This	group,	containing	skates	(Rajidae)	and	small	demersal	sharks,	inherited	most	(~90%)	of	the	biomass	of	the	original	“skate”	group	in	the	NBC	model,	reflecting	the	high	relative	abundance	of	skates.	Its	biomass	density	(B)	of	0.3	t/km2	was	derived	from	the	figure	for	the	original	group	minus	the	biomasses	of	the	large	sharks	(salmon,	blue	and	large	demersal).	The	P/B	(0.32	yr-1)	and	Q/B	(1.24	yr-1)	values	for	this	group	were	directly	inherited	from	the	parent	“skate”	group,	as	was	the	proportion	of	this	group	in	sperm	whale	diet	and	most	of	the	“skate”	bycatch	in	the	groundfish	trawl	fishery	(some	was	allocated	to	large	demersal	sharks).	The	diet	composition	of	this	 group	 (mostly	 benthic	 invertebrates)	 was	 also	 derived	 from	 that	 of	 its	 parent	 group,	rescaled	to	account	for	the	removal	of	the	prey	consumed	by	the	large	sharks.	Large	demersal	sharks	This	 group	 is	 composed	 of	 large	 ambush	 predators	 and	 scavengers	 such	 as	 bluntnose	 sixgill	shark	 (Hexanchus	 griseus),	 broadnose	 sevengill	 shark	 (Notorynchus	 cepedianus)	 and	 Pacific	sleeper	 shark	 (Somniosus	 pacificus).	 These	 species	 take	 a	 large	 variety	 of	 prey	 from	 benthic	invertebrates	to	large	squid	and	pinnipeds,	both	demersally	and	in	the	water	column.	Ecopath	parameter	 values	 (B,	P/B	and	Q/B)	 for	 large	demersal	 sharks	were	 sourced	 from	 the	EwE	model	of	Southeast	Alaska	built	by	Guénette	 (2005),	with	the	Q/B	value	adjusted	slightly	downward	 to	 accord	 better	 with	 those	 for	 other	 elasmobranchs	 in	 the	 HG	 model.	 The	 diet	composition	for	this	group	was	derived	from	the	same	source	as	the	Ecopath	parameter	values.	Salmon	sharks	(Lamna	ditropis)	and	Blue	sharks	(Prionace	glauca)	Each	of	these	large	pelagic	predators	was	allocated	its	own	functional	group	in	the	HG	model.	In	the	summer,	both	of	 these	species	 form	 large	aggregations	 in	Queen	Charlotte	Sound,	where	they	apparently	feed	on	salmon	bound	for	southern	spawning	grounds	(Williams	et	al.	2010).	Biomass	 densities	 (B)	 for	 both	 species	 were	 estimated	 as	 0.02	 t/km2	 based	 on	 pelagic	 shark	survey	results	from	the	southern	portion	of	the	model	area	published	by	Williams	et	al.	(2010).	The	 latter	 estimated	 that	 the	 pelagic	 shark	 assemblage	 surveyed	 consists	 of	 roughly	 equal	proportions	of	salmon	and	blue	sharks,	and	we	assumed	based	on	the	survey	results	that	half	of	these	individuals	occupy	the	model	area.	P/B	and	Q/B	values	for	salmon	and	blue	sharks	were	derived	from	Preikshot	(2005).	P/B	values	for	these	groups	were	estimated	at	0.20	yr-1	and	0.17	yr-1,	 respectively,	 while	 the	 corresponding	 Q/B	 estimates	 were	 1.20	 yr-1	 and	 0.80	 yr-1.	 Diet	composition	data	for	salmon	and	blue	sharks	were	taken	from	Hulbert	et	al.	(2005)	and	Nakano	and	Seki	(2003),	respectively,	with	the	proportion	of	total	annual	feeding	occurring	in	the	model	2016  Fisheries Centre Research Reports 24(2) 		32	area	(25%)	estimated	based	on	migration	data	published	by	(Weng	et	al.	2008).	The	remaining	75%	of	their	diet	was	classed	as	“import”	in	the	model’s	diet	composition	matrix.	Crabs	As	in	the	NBC	model,	there	are	two	size-based	functional	groups	of	crabs	in	the	HG	model:	large	crabs	(carapace	length	>120	mm)	and	small	crabs	(carapace	length	<120	mm).		Large	species	include	the	Dungeness	crab	(Cancer	magister),	red	rock	crab	(C.	productus),	tanner	crab	 (Chionoecetes	 bairdi),	 and	 king	 crab	 (Paralithodes	 spp.).	 The	 Dungeness	 crab	 is	commercially	the	most	important	crab	species	and	ranks	2nd	in	terms	of	landed	value	among	all	invertebrate	 fisheries	 on	 the	west	 coast	 of	 Canada	 (DFO	 2000a).	Management	 of	 Dungeness	crab,	understood	to	be	“fully	exploited”,	 is	mostly	based	on	size	and	sex	(DFO	2000a).	Smaller	species,	such	as	kelp	crab	(Pugettia	producta),	and	the	younger	life	stages	of	the	large	species,	were	 categorized	 as	 “small	 crabs.”	 Biomass	 and	 other	 Ecopath	 parameters	 for	 these	 groups	were	kept	unchanged	from	the	NBC	model	(Williams	et	al.	2010).	Commercial	Shrimp	This	group	 represents	all	 seven	commercial	 shrimp	and	prawn	species	 inhabiting	 the	coast	of	BC,	all	belonging	 to	 the	 family	Pandalidae.	These	are	smooth	shrimp	(Pandalus	 jordani),	 spiny	shrimp	(P.	borealis),	pink	shrimp	(P.	montagui),	coonstripe	shrimp	(P.	danae),	humpback	shrimp	(P.	 hypsinotus),	 sidestripe	 shrimp	 (Pandalopsis	 dispar)	 and	 prawn	 (P.	 platycerus)	 (Ainsworth	2006).	Trawl	nets	and	 traps	are	 the	 two	common	gears	 for	 commercial	harvesting	of	 shrimps	and	prawns	(DFO	2016).	The	 parameters	 of	 this	 group	 have	 not	 been	 changed	 in	 the	 present	 HG	 model.	 Detritus,	euphausiids	and	copepods	together	constitute	nearly	90%	of	its	diet.	Epifaunal	and	Infaunal	(carnivorous	and	detritivorous)	invertebrates	These	groups	were	 kept	unchanged	 from	 the	NBC	model	 (Ainsworth	2006).	 They	 represent	 a	great	diversity	of	invertebrate	species	that	were	not	assigned	distinct	groups	in	the	current	HG	model,	 although	 the	 representation	of	 epifaunal	 invertebrates	will	 be	 improved	 in	 the	 future	(Part	 B).	 Epifaunal	 invertebrates	 include	 members	 of	 Echinodermata	 (sea	 urchins,	 sea	 stars,	brittle	 stars,	 crinoids),	 Mollusca	 (gastropods,	 chitons,	 bivalves),	 Cnidaria	 (sea	 pens)	 and	Arthropoda	 (barnacles,	 amphipods).	 The	 infaunal	 carnivorous	 invertebrate	 group	 is	 primarily	composed	of	annelids	 (nereids,	bloodworms	etc.).	 Infaunal	detritivorous	 invertebrates	 include	gastropods,	 bivalves,	 echinoderms	 (sea	 cucumbers),	 amphipods,	 various	 worm	 phyla	 etc.	(Ainsworth	2006).	Biomass	 of	 the	 epifaunal	 invertebrates	 was	 estimated	 by	 Ecopath	 using	 its	 mass-balance	assumption,	while	the	biomasses	of	both	of	the	infaunal	groups	were	derived	from	that	of	the	“benthic	 infauna”	group	in	the	HS	model	developed	by	Beattie	(2001).	The	latter	was	also	the	source	for	the	Q/B	values	for	all	three	groups	and	P/B	for	epifaunal	and	infaunal	detritivorous	invertebrates.	The	P/B	for	infaunal	carnivorous	invertebrates	was	based	on	the	functional	group	“Invertebrate	Benthos”	in	the	southern	BC	shelf	model	of	1996	(Ainsworth	2006).	A Revised EwE Model of Haida Gwaii 		 33	Carnivorous	jellyfish	The	moon	 jellyfish	 (Aurelia	 aurita)	 is	 the	most	 “widely	 recognized”	 jellyfish	 among	 a	 total	 of	over	200	species	 found	globally;	however,	Rhopilema	esculenta	 is	 the	most	harvested	species	(Common	 Jellyfish	 (Aurelia	 aurita)	 	 n.d;	 Lucas	 2011).	 The	 increasing	 abundance	 trends	 and	“outbreaks”	 of	 many	 jellyfish	 populations	 over	 the	 last	 few	 decades	 are	 thought	 to	 be	 an	consequence	of	 the	 loss	of	marine	biodiversity	caused	by	unsustainable	 fisheries	 (Duffy	2015;	Richardson	et	al.	2009).	Frequent	occurrence	of	jellyfish	(A.	aurita)	in	the	summer	off	southern	BC	coast	motivated	researchers	to	conduct	a	test	fishery	in	late	1984;	however,	the	final	market	product	did	not	impress	the	Asian	consumers,	mainly	because	of	its	“low	protein	content”,	and	therefore	the	jellyfish	fishery	was	not	considered	“viable”	(Sloan	&	Gunn	1985).	This	group	also	contains	large	predatory	species	such	as	the	lion’s	mane	(Cyanea	capillata)	and	fried	egg	jellyfish	(Phacellophora	camschatica).	Ecopath	parameters	for	the	group	are	the	same	as	in	the	NBC	model.	As	no	direct	fisheries	for	jellyfish	exist	in	BC,	all	landings	were	marked	as	“bycatch	and	discards”	from	groundfish	trawl	and	salmon	gillnet	fisheries	(Ainsworth	2006).	Euphausiids	and	Copepods	Euphausiids,	also	known	as	krill,	play	vital	role	in	the	food	web	of	the	west	coast	of	BC,	as	they	predominantly	consume	phytoplankton	and	are	preyed	upon	by	a	number	species,	ranging	from	forage	 fish	 to	 the	 largest	 animal	 on	 the	 planet,	 the	 blue	whale.	 Some	 species	 rely	 on	 krill	 so	heavily	 that	 its	 abundance	may	 determine	 their	 distribution	 (for	 example,	 the	 distribution	 of	Pacific	hake	and	the	habitat	“shift”	of	humpback	whales	towards	the	plenitude	of	krill	 (DFO)).	Species	such	as	Pacific	herring	and	hake	are	among	the	 leading	fish	predators	of	krill	 in	Haida	Gwaii	 waters,	 and	 the	 diet	 of	 the	 blue	 whale	 constitutes	 predominantly	 of	 krill	 (DFO	 n.d.).	Twenty	 species	 out	 of	 a	 global	 total	 of	 85	 are	 found	 off	 the	 BC	 coast,	 though	 most	 of	 the	population	is	dominated	by	only	a	few	species	such	as	Euphausia	pacifica,	Thysanoessa	spinifera	and	T.	longipes	(DFO	n.d.).	E.	pacifica	is	more	nutritious	that	the	Thysanoessa	species,	and	might	be	partitioned	in	future	versions	of	the	model.	A	vast	proportion	of	the	crustacean	zooplankton	in	 the	ocean	are	copepods	 (DFO	2011b).	As	mentioned	 in	Williams	et	al.	 (2010),	 the	copepod	group	in	the	NBC	and	HG	models	is	comprised	of	three	major	genera:	Pseudocalanus,	Oithona	and	Acartia.	These	organisms	provide	prey	for	forage	fish,	herring	and	the	larvae	and	juveniles	of	many	species.	We	have	not	made	changes	in	these	functional	groups	and	their	Ecopath	parameters	B	and	P/B,	from	 Ainsworth	 (2006),	 who	 used	 estimates	 from	 Beattie	 (2001).	 Q/B	 for	 both	 groups	 was	estimated	assuming	P/Q	=	0.3	(Ainsworth	2006).	Corals	and	sponges		Corals	(Cnidaria)	and	sponges	(Porifera)	are	found	in	shallow	as	well	as	deep	waters	around	the	world	 (Roberts	 et	 al.	 2006).	 There	are	about	80	 species	of	 corals	 and	250	 species	of	 sponges	present	 in	 the	 waters	 off	 the	 west	 coast	 of	 Canada;	 they	 provide	 nursery	 grounds	 for	 the	juveniles	of	many	species,	e.g.	 rockfish	(DFO	2010b).	The	group	“Corals	and	sponges”	and	the	associated	parameters	in	the	HG	model	were	not	changed	from	the	NBC	model.	In	this	model,	their	diet	consists	exclusively	of	detritus	(Ainsworth	2006).	Phytoplankton	and	Macrophytes	The	identity	and	parameters	of	these	groups	were	left	unchanged	from	the	NBC	model.		2016  Fisheries Centre Research Reports 24(2) 		34		Figure	8.	Diets	matrix	used	in	HG	model.	For	clarity,	the	prey	contribute	less	than	0.05	in	the	diet	matrix	were	omitted	in	the	picture.	 	A Revised EwE Model of Haida Gwaii 		 35	Addition	of	new	fisheries A	total	of	5	new	fisheries	(4	Haida	fisheries	and	1	hake	fishery)	were	added	to	the	existing	fleets	in	the	NBC	model	(Figure	9	and	Figure	10).	The	Haida	fisheries	target	salmon,	herring	spawn	on	kelp,	 clams	 and	 seaweed.	 Although	 other	 fisheries	 occur	 in	 the	 model	 area,	 these	 were	identified	to	be	of	specific	interest	for	the	policy	options	to	be	investigated.	Landing	estimates	for	 salmon,	 herring	 spawn-on-kelp	 (SOK),	 and	 seaweed	 are	 for	 Food,	 Social	 and	 Ceremonial	(FSC)	 fisheries.	 Clam	 landings	 included	 the	 commercial	 razor	 clam	 fishery.	 Haida	 landings	 are	small	relative	to	commercial	fisheries	except	for	razor	clam	fisheries	(Russ	Jones,	pers.	comm.).	Fisheries	included	in	the	HG	model	were	exploited	by	a	total	of	17	fleets	with	various	gear	types	such	as	bottom	trawl,	gillnet,	purse	seine,	longline,	trap,	troll	etc.	Figure	9.	Total	landings	for	each	fleet	included	in	the	HG	model	2016  Fisheries Centre Research Reports 24(2) 		36		Figure	10.	Total	landings	of	each	functional	group	in	the	model,	categorized	by	fleet.	Incorporation of spatial information Delineation	of	the	boundaries	of	the	Haida	Gwaii	Ecospace	map	For	 the	 spatial	 analysis,	 the	 total	 area	 (land	 and	 water)	under	 study	 was	 divided	 into	 square	 grid	 cells,	 each	representing	an	area	of	16	km2.	The	land-water	boundary	was	mapped	based	on	shape	files	for	the	BC	coastline.	We	used	 maps	 developed	 using	 the	 BC	 Albers	 projection	because	 compared	 to	 the	 Mercator	 projection,	 this	approach	 preserves	 the	 distance	 from	 one	 point	 to	another	 on	 the	 map.	 Maintaining	 correct	 distances	 is	important	from	the	perspective	of	both	species	movement	and	sailing	costs	for	fishers.		Previously,	three	Ecospace	models	have	been	built	for	the	NBC	ecosystem.	Beattie	(2001)	built	a	model	with	a	large	extent	of	70,000	km2	and	developed	the	 Ecoseed	 routine	 for	 finding	 optimal	 location	 and	 size	 of	 MPAs.	 Salomon	 et	 al.	 (2002)	developed	an	Ecospace	model	 for	a	1,600	km2	area	within	 the	Gwaii	Haanas	National	Marine	Conservation	 Area	 Reserve,	 with	 a	 cell	 area	 of	 4	 km2.	 They	 used	 the	 model	 to	 explore	 the	Dimensions	Number	of	rows	 97	Number	of	columns	 60	Spatial	reference	Top-left	latitude	 54.3	N	Top-left	longitude	 134.1	W	Cell	side	length	(km)	 4.0	Cell	 size	 (decimal	degrees)	0.036	A Revised EwE Model of Haida Gwaii 		 37	optimal	size	of	MPAs	and	adjoining	buffer	areas	and	found	that	MPA	performance	was	related	to	concentration	of	 fishing	effort	 in	spillover	zones	around	the	MPA	boundaries,	which	drains	the	biomass	of	mobile	species	from	the	MPAs.	This	meant	that	 larger	MPAs	performed	better	due	to	their	higher	ratio	of	area	to	perimeter.	The	boundaries	of	the	HG	Ecospace	model	are	based	on	the	spatial	extent	of	the	HG	EwE	model	and	 enclose	 a	 total	 area	 of	 81,008	 km2.	 Essentially,	 they	 include	 the	 inshore	 and	 offshore	distribution	of	many	key	species	in	the	model.	However,	the	migrations	of	salmon,	large	whales	and	some	sharks	extend	far	outside	the	model	area,	and	therefore	the	model	faces	limitations	when	attempting	to	explain	the	dynamics	of	such	species.	With	respect	to	herring,	the	modelled	area	includes	the	distributions	of	the	HG	and	HG	2W	as	well	as	most	of	the	PRD	and	CC	stocks.	A	high-resolution	model	enables	spatial	analyses	at	a	very	fine	scale,	especially	for	species	with	small	ranges.	The	choice	of	resolution	(i.e.	Ecospace	cell	size)	is	based	on	the	research	question,	the	range	of	key	species	in	the	system,	and	the	trade-off	between	the	resolution	of	spatial	data	available	and	computation	time.	We	chose	a	4	x	4	km	resolution	for	 the	map	(Figure	11):	our	model	includes	species	with	very	large	ranges	and	4	x	4	km	was	the	smallest	resolution	that	we	could	 adopt	 before	 noticing	 a	 large	 increase	 in	 computation	 time	 for	 running	 one	 Ecospace	scenario.	 At	 a	 4	 x	 4	 km	resolution,	 we	 were	 able	 to	capture	 a	 fairly	 realistic	representation	 of	 the	coastline.	We	 failed,	 however,	to	 include	 a	 narrow	 (<4	 km	wide)	 but	 ecologically	important	 channel	 between	Moresby	 and	 Anthony	 Islands	in	 Gwaii	 Haanas	 (Anne	Salomon,	 SFU,	 pers.	 comm).	(This	 channel	 would	 be	manually	 drawn	 into	 the	map	to	allow	for	species	movement	between	the	East	and	West	of	Haida	Gwaii	 in	 the	 revised	HG	model).	 Skidegate	 Channel	between	 Graham	 and	Moresby	 Islands	 was	 also	 not	included	 in	 the	 original	 map	and	 will	 be	 entered	 manually	in	 the	 final	 version	 of	 the	model.		.	Figure	11.	The	Ecospace	base	map	for	the	HG	model.	The	 left	picture	shows	a	4X4	km	resolution	base	map	while	 the	panel	to	 the	right	shows	all	the	seven	habitat-types	used	in	the	model.	2016  Fisheries Centre Research Reports 24(2) 		38	Ecospace	habitat	capacity	Ecospace	 habitat	 capacity	 maps	 were	 designed	 for	 each	 functional	 group	 in	 the	 model.	 GIS	shape	 files	 for	 relative	 abundances	 of	 several	 species	 were	 obtained	 from	 the	 Haida	 Ocean	Technical	 Team	 (HOTT),	 and	 maps	 from	 the	 Haida	 Marine	 Traditional	 Ecological	 Knowledge	report	 (Council	 of	 the	 Haida	 Nation,	 2011).	 We	 explored	 two	 approaches	 for	 extracting	information	for	the	HG	map	area	from	the	shape	files:		(i)	We	 created	 an	 empty	map	 boundary	 for	 the	 HG	 area	 and	 cut	 shape	 files	 to	 this	 area	 to	produce	HG-specific	shape	files.	However,	we	found	that	for	several	species	the	shape	files	for	relative	abundance	did	not	encompass	the	full	area	of	the	HG	map	and	often	excluded	the	left	boundary.	For	this	reason,	we	decided	that	the	method	of	clipping	existing	shape	files	to	fit	our	area	was	not	appropriate.	(ii)	We	converted	all	shape	files	provided	into	ASCII	files	using	ESRI	ArcMap	(version	10.2).	We	wrote	programs	in	R-software	to	extract	the	HG	area	map	from	the	ASCII	files.	Parts	of	the	map	that	 had	 no	 information,	 usually	 the	 left	 boundaries,	 were	 assigned	 ‘zero’	 habitat	 capacity.	However,	 in	 the	 final	 review	 process,	 the	 habitat	 capacity	 maps	 generated	 were	 carefully	reviewed	 and	 when	 available,	 information	 from	 other	 sources	 was	 entered	 according	 to	suggestions	of	the	HOTT	and	other	experts.	Following	are	the	steps	we	used	to	generate	maps	for	our	Ecospace	model.	Conversion	of	shape	files	to	ASCII:	1. Add	the	shape	file	in	ArcGIS	2. We	 followed	 these	 steps	 to	 create	a	 raster	 file:	Arc	 Toolbox	à	 Conversion	 tools	àTo	Raster	à	 Polygon	 to	Raster.	 Cell	 size	was	 set	 to	4000	 for	 every	 time	a	 raster	 file	was	created.	This	was	because	in	our	Ecospace	map	the	cell	size	is	4	x	4	km.		3. Then	under	the	same	section:	Conversion	tools	àFrom	Raster	à	Raster	to	ASCII.	After	conversion	to	ASCII	format,	code	developed	using	R	was	used	to	select	the	area	from	the	ASCII	file	that	was	relevant	to	the	Haida	Gwaii	map	area.	Cases	a,	b,	c,	and	d	in	Figure	12	show	the	examples	where	the	lower	left	margins	of	the	ASCII	file	were	different	and	we	developed	an	algorithm	to	extract	relevant	data	from	ASCII	files.		Figure	12.	Extraction	of	HG	data	for	Ecospace	from	GIS	files.	Blue	 shows	ASCII	 file	 boundary,	 orange	 shows	HG	 area,	 grey	 indicates	 land	mass,	 and	 the	 area	 shaded	with	 lines	 shows	 the	relevant	data	 that	were	extracted	 from	 the	ASCII	 files.	 Panels	a,	 b,	 c,	 and	d	 indicate	 four	different	 types	of	 spatial	 extent	 for	which	the	original	shape	files	were	available.	A Revised EwE Model of Haida Gwaii 		 39	The	code	for	extracting	the	relevant	data	(shown	as	shaded	area	in	Figure	12)	and	develop	the	habitat	capacity	map	for	the	Haida	Gwaii	Ecospace	map	area	is	provided	in	Appendix	D.	For	40	out	 of	 56	 functional	 groups,	 the	habitat	 capacity	 layers	were	mapped	based	on	 available	GIS	maps.	 In	 some	 cases,	 the	 shape	 files	 for	 several	 species	 were	merged	 to	 create	 the	 habitat	capacity	 layer	 for	 a	 functional	 group.	 For	 other	 species,	 the	 habitat	 capacity	 maps	 were	developed	based	on	literature	reviews	of	occurrence	in	different	depth	ranges	and	geographical	areas.	 The	 Ecospace	 habitat	 capacity	 maps	 for	 salmon	 and	 blue	 sharks	 were	 derived	 from	sightings	data	collected	on	a	line	transect	survey	by	Williams	et	al.	(2010).	For	cetacean	groups,	the	maps	derived	 from	shape	 files	were	modified	based	on	critical	habitat	areas	predicted	by	Gregr	and	Trites	(2001)	using	sightings	and	whaling	catch	data,	as	well	as	recent	sightings	and	historical	whaling	catches	recorded	in	Ford	(2014).	The	habitat	capacity	maps	for	seals	and	sea	lions	were	modified	based	on	Ford	(2014)	and	personal	communications	from	Trites	(2014).	Table	1.	Shape	files	obtained	from	the	Haida	Ocean	Technical	Team	(HOTT)	SN	 Functional	groups	 Name	 of	 original	GIS	shape	file	ASCII	file	name	 Comments	1	 Sea	Otters	 DFO_EBSAs	 rastert_nichol_1	 In	this	file,	sea	otters	were	not	found	around	the	HG	island,	 and	 seas	 otter	 distribution	 was	 sketched	based	on	 recent	 sightings	 reported	by	HG	 residents.	In	addition,	low	capacity	(0.2)	was	allowed	in	inshore	areas.	2	 Gray	whales	 DFO_EBSAs	 rastert_ford_gr2	 	3	 Humpback	whales	 DFO_EBSAs	 rastert_humpbac1	 Improved	 with	 sightings	 and	 whaling	 catch	 records	(Ford	2014).	4	 Minke	whales	 BCMCA	 	 Sightings	records	(Ford	2014)	5	 Blue	whales	 DFO_EBSAs	 rastert_ford_bl1	 Improved	 with	 sightings	 and	 whaling	 catch	 records	(Ford	2014).	6	 Fin	whales	 DFO_EBSAs	 rastert_ford_fi1	 Improved	 with	 sightings	 and	 whaling	 catch	 records	(Ford	2014).	7	 Sei	whales	 DFO_EBSAs	 rastert_ford_se1	 Improved	 with	 sightings	 and	 whaling	 catch	 records	(Ford	2014).	8	 Sperm	whales	 DFO_EBSAs	 rastert_ford_sp1	 Improved	 with	 sightings	 and	 whaling	 catch	 records	(Ford	2014).	9	 Resident	orcas	 DFO_EBSAs	 rastert_residen1	 Improved	with	sightings	records	(Ford	2014).	10	 Transient	orcas	 	 	 Sightings	records	(Ford	2014)	11	 Small	odontocetes	 	 	 Dolphin	habitat	12	 Seals	 DFO_EBSAs	 rastert_olesiuk1	 Improved	with	sightings	records	(Ford	2014).	13	 Sea	lions	 DFO_EBSAs	 rastert_steller1	 Improved	with	sightings	records	(Ford	2014).	14	 Seabirds	 PNCIMA	 rastert_birds_p1	 		15	 Transient	salmon	 Salmon	 rastert_scea-tr1	 GIS	data	sparse	and	habitat	capacity	to	all	regions	up	to	1000m	and	low	capacity	(0.2)	into	deeper	areas.	16	 Coho	salmon	 Salmon	 rastert_scea-tr2	 GIS	data	sparse	and	habitat	capacity	to	all	regions	up	to	100m	17	 Chinook	salmon	 Salmon	 rastert_scea-tr3	 GIS	data	sparse	and	habitat	capacity	to	all	regions	up	to	100m	18	 Small	squid	 	 	 All	areas	from	10	to	1000m	19	 Large	squid	 	 	 All	areas	deeper	than	10m	20	 Ratfish	 	 	 All	areas,	with	very	low	capacity	in	deeper	areas	2016  Fisheries Centre Research Reports 24(2) 		40	SN	 Functional	groups	 Name	 of	 original	GIS	shape	file	ASCII	file	name	 Comments	21	 Dogfish	 DFO_Catch_Data_Aggregated	rastert_sched2_1	 GIS	data	sparse	and	habitat	capacity	to	all	areas	with	very	low	capacity	in	deeper	areas	22	 Pollock	 DFO_EBSAs	 rastert_species2	 GIS	data	sparse	and	habitat	capacity	to	all	regions	up	to	500m	23	 Forage	fish	 DFO_Catch_Data_2012_Updates	rastert_dfo_bc_1	 GIS	data	sparse	and	habitat	capacity	to	all	regions	up	to	1000m	with	very	low	capacity	in	deeper	areas	24	 Hake	 DFO_EBSAs	 rastert_cooke_h1	 GIS	data	sparse	and	habitat	capacity	to	all	regions	up	to	1000m	with	very	low	capacity	in	deeper	areas	25	 Eulachon	 DFO_EBSAs	 rastert_hay_eul1	 Habitat	 capacity	 based	 on	 GIS	 data	 and	 updated	based	on	pers.	comm.	from	John	Kelson	26	 Juvenile	herring	 BCMCA	 rastert_bcmca_e1	 GIS	data	updated	with	low	capacity	in	deeper	areas	27	 Adult	herring	 DFO_EBSAs	 rastert_herring1	 GIS	 data	 updated	with	 low	 capacity	 in	 deeper	 areas	(0.9	 in	 areas	 up	 to	 200m,	 0.5	 in	 areas	 from	 200	 to	500m,	0.1	in	deeper	areas).		28	 Pacific	 Ocean	Perch		 	 Areas	from	10	to	200m	29	 Inshore	rockfish	 DFO_Catch_Data_Aggregated	rastert_zn_1	 GIS	data	sparse	and	habitat	capacity	in	inshore	areas	30	 Piscivorous	rockfish	DFO_EBSAs	 rastert_species7	 GIS	 data	 updated	with	 low	 capacity	 in	 deeper	 areas	(0.5	in	100	to	500m)		31	 Planktivorous	rockfish	DFO_Catch_Data_Aggregated	rastert_zn_1	 GIS	 data	 updated	with	 low	 capacity	 in	 deeper	 areas	(0.5	in	100	to	500m)	32	 Arrowtooth	flounder		 	 Areas	from	10	to	200m	33	 Flatfish	 DFO_EBSAs	 rastert_species6	 GIS	data	updated	with	small	capacity	in	deeper	areas	34	 Juvenile	halibut	 	 	 Areas	up	to	100m	35	 Adult	halibut	 DFO_EBSAs	 rastert_species3	 All	areas	up	to	500m	36	 Pacific	cod	 DFO_EBSAs	 rastert_species1	 GIS	data	updated	with	small	 (0.1)	capacity	 in	deeper	areas	37	 Sablefish	 DFO_Catch_Data_Aggregated	rastert_sablefish_2	 GIS	data	updated	with	small	 (0.1)	capacity	 in	deeper	areas,	 modified	 to	 eliminate	 spurious	 biomass	concentration	 in	 northern	Hecate	 Strait	 and	 eastern	Dixon	Entrance	38	 Lingcod	 DFO_EBSAs	 rastert-species5	 All	areas	up	to	200m	and	0.5	from	200	to	500m	39	 Shallow-water	benthic	fish		 	 All	areas	up	to	100m	40	 Small	 demersal	elasmobranchs		 	 All	areas	up	to	200m	and	0.5	from	200	to	500m	41	 Large	 demersal	sharks		 	 Everywhere	42	 Salmon	sharks	 	 	 All	areas	except	inshore	areas	43	 Blue	sharks	 	 	 All	areas	except	inshore	areas	44	 Large	crabs	 DFO_Catch_Data_Aggregated	rastert_crab_001.txt	 GIS	data	updated	with	0.5	in	areas	up	to	100m		45	 Small	crabs	 DFO_Catch_Data_2012_Updates	rastert_dfo_bc_2	 GIS	 data	was	 only	 for	 red	 rock	 crab	 and	 the	 habitat	capacity	was	0.5	in	areas	up	to	200m,	0.4	in	coral	and	sponge	areas,	0.3	 in	areas	200	to	500m,	0.2	 in	areas	500	to	1000m,	and	0.1	in	deeper	areas	46	 Commercial	shrimp	 DFO_EBSAs	 rastert_phillip3	 		A Revised EwE Model of Haida Gwaii 		 41	SN	 Functional	groups	 Name	 of	 original	GIS	shape	file	ASCII	file	name	 Comments	47	 Epifaunal	invertebrates	DFO_Catch_Data_2012_Updates	rastert_dfo_bc_5	 GIS	 data	 was	 only	 for	 octopus	 and	 habitat	 capacity	areas	up	to	10m	and	smaller	capacity	in	deeper	areas	48	 Infaunal	carnivorous	invertebrates		 	 In	 areas	 up	 to	 10m	 and	 smaller	 capacity	 in	 deeper	areas	49	 Infaunal	detritivorous	invertebrates	DFO_Catch_Data_Aggregated	 and		DFO_EBSAs	rastert_seacuc_1	 and	rastert_hand_se1	GIS	 data	 was	 only	 for	 sea	 cucumber	 and	 habitat	capacity	extended	to	all	regions	50	 Carnivorous	jellyfish		 	 Everywhere	51	 Euphausiids	 DFO_Catch_Data_Aggregated	rastert_dfo_bc_3	 GIS	 data	 sparse	 and	 distribution	 extended	 up	 to	500m	and	lower	capacity	in	deeper	areas	52	 Copepods	 	 	 Everywhere	53	 Corals	and	sponges	 PNCIMA	 full_coral_ascii	 		54	 Macrophytes	 BCMCA	 rastert_bcmca_e3	rastert_bcmca_e4		pdf	map	for	bull	kelp	Macrophyte	distribution	was	based	on	the	two	shape	files	 for	 giant	 kelp	 and	 eelgrass,	 and	 a	 PDF	map	 for	bull	kelp	55	 Phytoplankton	 BCMCA	 rastert_chlorop1	 This	was	not	used		 Ecospace habitat capacity maps 		 	 	 		 	 	 	2016  Fisheries Centre Research Reports 24(2) 		42			 	 	 		 	 	 		 	 	 	A Revised EwE Model of Haida Gwaii 		 43			 	 	 		 	 	 		 	 	 	2016  Fisheries Centre Research Reports 24(2) 		44			 	 	 		 	 	 		 	 	 	A Revised EwE Model of Haida Gwaii 		 45			 	 	 		 	 	 		 	 	 	2016  Fisheries Centre Research Reports 24(2) 		46			 	 	 		 	 	 	Figure	13.	Habitat	capacity	map	Spatial	distribution	of	fishing	fleets	Maps	 of	 fishing	 activities	 for	 several	 fleets	were	 available	 from	Haida	Ocean	 Technical	 Team	(HOTT).	We	used	an	inverse	of	the	effort	maps	obtained	and	provided	these	maps	as	input	for	sailing	cost	distribution.	The	resulting	pattern	was	that	effort	was	directed	to	areas	where	effort	was	observed	or	believed	to	occur	based	on	the	original	maps.		Table	2.	The	ArcGIS	files	used	for	mapping	the	spatial	distribution	of	fleets	in	the	model	sl	no	Fleet	name	 ASCII	file	name	 Source	of	original	shape	file	 Comments	1	 Groundfish	trawl	 rastert_ground_1	 DFO_Catch_Data_Aggregated	 multispecies	2	 Sablefish	fisheries	rastert-sablefish_1,	rastert-sablefish_2	DFO_Catch_Data_Aggregated	 Longlines	and	traps	3	 Herring	gillnet	 rastert_feature1	 Herring_Catch_Data	 	4	 Ground	H+L	 rastert_sched2_1	 DFO_Catch_Data_Aggregated	 	5	 Salmon	gillnet	 rastert_scea_gn1	 Salmon	 PNCIMA	6	 Crab	trap	 rastert_crab_001.txt,	rastert_dfo_bc_12	DFO_Catch_Data_Aggregated,	DFO_Catch_Data_2012_Updates	Data	for	crab	and	king	crab	7	 Shrimp	 /	 prawn	trap	rastert_dfo_bc_6	 DFO_Catch_Data_2012_Updates	 docks		  rastert_dfo_bc_10	 DFO_Catch_Data_2012_Updates	 humpback	shrimp		  rastert_dfo_bc_11	 DFO_Catch_Data_2012_Updates	 prawn		  rastert_pinkshm1	 DFO_Catch_Data_Aggregated	 pink	shrimp		  rastert_prawn_02	 DFO_Catch_Data_Aggregated	 prawn	traps		  rastert_sidestr1	 DFO_Catch_Data_Aggregated	 sidestripe	shrimp	8	 Other	Inv.	 rastert_geoduck1	 DFO_Catch_Data_Aggregated	 geoduck		  rastert_dfo_bc_8	 DFO_Catch_Data_2012_Updates	 green	urchin,	DFO	A Revised EwE Model of Haida Gwaii 		 47	sl	no	Fleet	name	 ASCII	file	name	 Source	of	original	shape	file	 Comments		  rastert_dfo_bc_9	 DFO_Catch_Data_2012_Updates	 red	urchin,	DFO	9	 Halibut	H+L	 rastert_feature3	 DFO_Catch_Data_2012_Updates	 	10	 Salmon	troll	 rastert_scea_tr1	 Salmon	 	11	 Salmon	seine	 rastert_scea_sn1	 Salmon	 	12	 Salmon	 troll	freezer	rastert_scea_tr1	 Salmon	 	13	 Herring	seine	 rastert_feature2	 Herring_Catch_Data	 	14	 Shrimp	trawl	 rastert_shrimp_1	 DFO_Catch_Data_2012_Updates	 	15	 Longline	 rastert_hay_eul1	 DFO_EBSAs	 	16	 Recreational	 rastert_sched2_1	 DFO_Catch_Data_Aggregated	 	  rastert_bcmca_h1	 BCMCA	 	  rastert_bcmca_h2	 BCMCA	 	  rastert_bcmca_h3	 BCMCA	 	  rastert_bcmca_h4	 BCMCA	 	17	 Hake	 rastert_cooke_h1	 DFO_EBSAs	 	18	 HG_salmon	 pdf	map	 Haida	 Marine	 Traditional	Knowledge	Study		19	 HG_herring_SOK	 pdf	map	 Haida	 Marine	 Traditional	Knowledge	Study		20	 HG_Clam	 pdf	map	 Haida	 Marine	 Traditional	Knowledge	Study		21	 HG_Seaweed	 pdf	map	 Haida	 Marine	 Traditional	Knowledge	Study			 	 	 	 	2016  Fisheries Centre Research Reports 24(2) 		48	Analyses	of	marine	protected	areas	We	explored	the	effect	of	marine	protected	areas	under	different	scenarios	of	fishing	effort	in	the	 region.	 MPA	 boundaries	 (Figure	 14)	 were	 developed	 based	 on	 the	 following	 two	classifications	obtained	from	PNCIMA.	1. Low	 target	 and	high	 clumping:	 The	MPAs	were	 designed	 to	protect	 10%	 of	representative	 features	and	 20%	 of	 special	features.	 The	 Marxan	analysis	 minimised	 the	total	 area	 closed	 ‘while	aiming	 for	 large-sized	clumps’	 (which	 was	achieved	 by	 setting	 the	Boundary	 Length	Modifier	 parameter	 in	Marxan	to	2500).	2. High	 target	 and	high	 clumping:	 The	MPAs	were	 designed	 to	protect	 40	 to	 50%	 of	representative	 and	special	 features.	 Here	again,	 the	 BLM	parameter	 was	 set	 to	2500.		We	 chose	 to	 use	 the	high	clumping	scenarios	 from	the	Marxan	analysis	because	 for	 the	medium	and	 low	clumping	scenarios,	the	results	 included	many	MPAs,	some	of	which	were	smaller	than	4km2	and	it	was	not	possible	to	include	these	within	the	Ecospace	maps.				  Figure	 14.	 MPA	 and	 spillover	 boundairies	 in	 Ecospace	 based	 on	 two-high	 clumping	 MPA	options.		MPAs	are	shown	in	blue,	cells	near	the	MPA	boundaries	are	spillover	regions	shown	in	orange.	Figure	a	is	for	low-target	MPAs	for	smaller	closed	area	and	Figure	b	is	for	high-target	MPAs	for	larger	closed	area.	A Revised EwE Model of Haida Gwaii 		 49	PART B: Another improvement to the HG ecosystem model Pacific herring One	of	the	premises	for	the	model	improvements	is	to	create	a	minimum	of	three	age	stanzas	for	each	of	the	four	Pacific	herring	stocks	falling	under	the	spatial	extent	of	the	modelling	area.	As	per	 the	DFO’s	management	 scheme,	 these	 stocks	are	HG	 (Haida	Gwaii),	HG	minor	 (Area	2	West),	 PRD	 (Prince	 Rupert),	 and	 CC	 (Central	 Coast).	 The	 three	 stanzas	will	 correspond	 to	 the	following	herring	age/size	classes: 1. Young	of	year	herring	 (ages:	0-12	months,	 fork	 length:	≤	10	cm),	which	spend	most	of	their	time	in	nearshore	waters,	2. Immature	 herring	 (ages:	 12-36	 months,	 fork	 length:	 10-20	 cm),	 which	 mostly	 stay	offshore	and	join	the	spawning	stock	at	the	end	of	the	stage,	and	3. Mature	 herring	 (ages:	 36+	 months,	 fork	 length:	 >20	 cm),	 the	 leading	 stanza,	 which	undergoes	 seasonal	 migration	 between	 offshore	 and	 nearshore	 waters	 and	 supports	most	of	the	herring	fisheries	such	as	roe	herring,	food	and	bait,	and	spawn-on-kelp.		Parameterization of Pacific herring For	 an	 age-structured	 functional	 group,	 Ecopath	 requires	 total	 mortality	 (Z)	 and	 trophic	interactions	 for	 each	 of	 the	 constituent	 age	 classes	 (stanzas)	 of	 the	 group.	 B	 and	 Q/B	 are	required	for	the	leading	stanza	only.	The	 biomasses	 for	 all	 four	 stocks	 of	 Pacific	 herring	 under	 study	 were	 obtained	 from	 DFO’s	assessment	 data	 generated	 using	 an	 integrated	 statistical	 catch-at-age	 model	 (DFO	 2015c).	Presently,	 the	 assessment	model	 (AM)	 is	 fitted	 to	 “commercial	 catch,	 proportions-at-age	 and	fishery-independent	 survey	 index”	 under	 two	management	 procedures:	 (1)	 the	model	 (AM1)	was	allowed	to	estimate	the	“spawn	survey	scaling	parameter”	q,	in	contrast	to	(2)	in	which	the	model	(AM2)	was	run	with	fixed	q.	The	management	procedure	based	on	the	AM1	approach	is	referred	to	as	the	“current”	management	procedure.	In	this	procedure,	the	herring	biomass	cut-off	 at	 which	 fisheries	may	 open	 is	 expressed	 as	 a	 proportion	 (0.25)	 of	 the	model’s	 unfished	biomass	 estimates.	 On	 the	 other	 hand,	 the	 management	 procedure	 relying	 on	 the	 AM2	approach	is	referred	to	as	the	“historical”	management	procedure,	in	which	the	cut-off	biomass	is	“fixed”	and	therefore,	independent	of	the	model’s	estimates	of	unfished	biomass.		The	assessment	models	consider	 recruitment	at	age	2	and	estimate	a	number	of	parameters,	including	 abundance	 dynamics	 and	 fishing	 mortality	 (F)	 for	 ages	 ≥	 2,	 total	 biomass	 (bt)	 and	spawning	 stock	 biomass	 (sbt)	 separately	 for	 each	 stock.	 The	models	 also	 estimate	 the	 time-varying	 natural	mortality	 (M),	which	 is	 assumed	 to	 be	 the	 same	 across	 all	 ages	 ≥	 2,	 for	 each	stock	in	a	given	year.		Parameterization	of	all	four	herring	stocks	in	the	revised	ecosystem	model	is	based	on	the	data	obtained	 from	DFO’s	 current	management	procedure	 (with	 the	exception	of	Q/B	and	 trophic	interactions	in	the	model	diet	matrix).	2016  Fisheries Centre Research Reports 24(2) 		50	Biomass Biomass	of	mature	herring	in	the	leading	stanzas	(ages	>	3)	for	all	four	stocks	were	estimated	as	a	sum	of	the	products	of	abundance	at	age	(Na)	and	weight	at	age	(Wa)	across	ages	3	to	10	for	the	respective	stocks,	Equation	(	8	):		 !!,![!",!!,!"#,!!] = !!,! ∗!!,!!"!!! 	 (	8	)		Total mortality (Z) The	assessment	model	provides	estimates	of	age-specific	F	for	all	stocks	but	does	not	explicitly	provide	F	for	ages	3+	that	are	required	for	the	HG	model.	We	calculated	F	for	age	3	using	the	equations	(	9	)	and	(	10	)	below:		 !!,![!",!!,!"#,!!] = 1− exp −!!,! ∗ !!,!;    !"# !"#$ 2 !" 10	 (	9	)			 !!!,   ![!",!!,!"#,!!] = !!,!!"!!!!!!,! 	 (	10	)		Since	in	the	assessment	model,	M	is	the	same	across	all	ages,	we	have	obtained	M	directly	from	the	assessment	model	and	estimated	Z	for	ages	3+	as	in	Equation	(	11	):		 !!!,   ![!",!!,!"#,!!] = !! + !!!,!	 (	11	)		Z	 for	age	2	was	calculated	as	the	sum	of	Fage	 2	and	M	for	all	 the	stocks.	As	mentioned	before,	DFO’s	assessment	models	use	 recruitment	at	age	2,	 and	 therefore	we	have	not	obtained	any	information	about	Z	for	the	youngest	herring	stanza.	We	used	a	value	Z	=1.23	from	a	study	on	cohort	analysis	in	the	eastern	Bering	Sea	(Wespestad	1982)	and	assumed	this	value	to	apply	to	all	stocks.	Catch DFO’s	assessment	model	uses	the	catches	obtained	from	commercial	seine	roe	fisheries,	 food	and	bait/special	use	fisheries	and	commercial	gillnet	roe	fisheries.	Though	most	of	the	catch	is	of	fish	aged	3	and	higher,	we	have	not	obtained	direct	estimates	of	age-specific	catches	for	all	fisheries	gears	across	all	stocks.		We	have	calculated	those	values	as:	1. Using	the	catch	Equation	(	9	),	we	calculated	age	2	and	age	3+	catches	and	determined	the	proportion	of	the	two	groups	in	total	estimated	catch,	2. Using	 this	 proportion,	 we	 have	 extracted	 age	 2	 and	 age	 3+	 catches	 from	 observed	catches	for	all	three	gears	used	in	the	assessment	model	across	all	stocks.	In	 the	 original	 HG	 ecosystem	model,	 two	 gears,	 namely	 herring	 gillnet	 and	 herring	 seine,	were	apportioned	the	entire	commercial	herring	catch.	To	match	the	gear-wise	commercial	A Revised EwE Model of Haida Gwaii 		 51	landings	 between	 the	 DFO’s	 assessment	 model	 and	 the	 HG	 ecosystem	 model,	 we	 have	included	 food	 and	 bait/special	 use	 fisheries	 in	 the	 revised	 HG	model.	 However,	 as	 there	were	no	commercial	catches	recorded	under	food	and	bait/special	use	in	2000,	the	fishery	act	as	a	placeholder	for	the	landings	in	subsequent	years.		We	 have	 also	 updated	 the	 adult	 (age	 3+)	 herring	 mortality	 associated	 with	 spawn-on-kelp	fisheries	 for	 each	of	 the	 stocks	 based	on	 the	 information	 available	 on	 the	DFO	website	 (DFO	n.d).		The	estimated	biomasses	and	catches	 for	each	stock	and	stanza	 in	 the	model	are	provided	 in	Table	3.	Table	3.	Age-structured	parameterization	of	Pacific	herring	across	the	four	stocks	in	the	revised	HG	model	(year	2000).		All	the	values	in	bold	were	estimated	by	Ecopath.	A	couple	of	values	for	Z	for	PRD	and	CC	stocks	were	slightly	adjusted	during	balancing	(originals	are	provided	in	the	parentheses)	Pacific	herring	 B	(t.km-2)	 Z	(/	year)	 Q/B	(year)	 	Herring	gillnet	(t.km-2)	Herring	seine	(t.km-2	)	HG	herring	 	     Age	0-1	year	 0.05	 1.23	 21.43	 	 	Age	1-3	years	 0.47	 1.00	 9.20	 	 0.0011	Age	3+	years	 0.31	 1.06	 5.84	 	 0.0207	HG_minor	herring	 	     Age	0-1	year	 0.00	 1.23	 23.67	 	 	Age	1-3	years	 0.00	 0.45	 9.56	 	 	Age	3+	years	 0.01	 0.45	 5.84	 	 	PRD	herring	 	     Age	0-1	year	 0.02	 1.23	 22.31	 	 	Age	1-3	years	 0.32	 0.63	(0.43)	 9.18	 0.0003	 0.0001	Age	3+	years	 0.47	 0.74	(0.54)	 5.84	 	 0.0376	 0.0152	CC	herring	 	     Age	0-1	year	 0.03	 1.23	 22.87	 	 	Age	1-3	years	 0.46	 0.52	(0.42)	 9.30	 0.0001	 0.0004	Age	3+	years	 0.98	 0.60	(0.51)	 5.84	 0.0119	 0.0785	Diet An	extensive	and	detailed	dietary	preference	analysis	has	been	conducted	in	the	waters	off	the	BC	Central	Coast.	 In	 total,	approximately	1300	stomachs	 taken	 from	herring	 ranging	 in	 length	from	about	0.5	cm	to	over	30	cm	have	been	analyzed.	Great	care	was	taken	directed	to	obtain	as	wide	a	 geographical	 and	 seasonal	 coverage	of	 the	herring	diet	 as	possible.	As	a	 result,	we	have	 a	 data	 set,	 collected	 between	 2007	 and	 2015,	 of	 sufficient	 geographical	 (from	 Queen	Charlotte	 Strait	 to	 northern	 Hecate	 Strait)	 and	 seasonal	 (March	 to	 November)	 resolution.	 In	addition,	historical	data	sets	available	through	the	DFO	reports	will	be	used	to	supplement	our	dietary	 information.	 Although	 prey	 components	 were	 identified	 to	 the	 highest	 possible	taxonomic	resolution,	only	major	plankton	groups	will	be	considered	in	the	model	improvement	process.		2016  Fisheries Centre Research Reports 24(2) 		52	The	preliminary	analysis	showed	that,	contrary	to	what	was	shown	in	previous	model	versions,	euphausiids	 contribute	 only	modestly	 to	 the	 annual	 diet	 of	 Pacific	 herring	 off	 the	 BC	 Central	Coast.	 Only	 in	 a	 few	 locations	 may	 they	 be	 a	 dominant	 food	 item.	 Other	 groups,	 including	copepods,	 amphipods,	 decapods,	 mollusks	 and	 fish	 larvae	 also	 appeared	 to	 be	 important	components	of	the	Central	Coast	herring	diet.	The	revised	Pacific	herring	diet	proposed	for	the	improved	Haida	Gwaii	model	is	presented	in	Table	4.	Table	4.	Prey	composition	(by	weight)	of	the	Pacific	herring	in	the	Central	BC	Coast.	Prey	groups	 Age	1	year		 Age	1-3	years		 Age	3+	years		(SL	<10	cm)	 (SL	10-20	cm)	 (SL	>20	cm)	Copepods	 65	 20	 15	Euphausiids	 5	 38	 15	Amphipods	 2	 18	 25	Fish	(larvae)	 0	 5	 22	Others	(breakdown	below)	 28	 19	 23	decapods	 5	 5	 5	mollusks	 5	 5	 10	cladocerans	 18	 0	 0	chaetognaths,	siphonophores,	others	 0	 9	 8		In	summary,	herring	smaller	than	10	cm	long	mainly	consume	small	copepods	and	cladocerans.	The	latter	are	indicative	of	their	coastal	habitat.	Herring	between	10	and	20	cm	in	length	prey	largely	 upon	 euphausiids,	 copepods	 and	 amphipods,	 while	 older	 and	 larger	 fish	 feed	opportunistically	 on	 a	 variety	 of	 prey	 groups,	 including	 larval	 fish,	 without	 a	 particular	preference.	 According	 to	 Table	 1,	 the	minimum	 set	 of	 plankton	 groups	 in	 the	 revised	model	should	include,	besides	the	jellyfish	component	already	existing	in	the	model,	copepods	(already	exist),	 euphausiids	 (already	 exist),	 amphipods,	 fish	 larvae	 and	 other	 plankton	 (mostly	combination	 of	 decapods,	 mollusks,	 cladocerans,	 chaetognaths	 and	 siphonophores	 +	 other	small	 groups).	 The	 “other	 plankton”	 category	 could	 be	 further	 broken	 down	 into	 three	categories:	decapods,	mollusks,	cladocerans	and	other	plankton	 (Table	4).	 Ideally,	 it	would	be	advantageous	to	have	decapods	and	mollusks	as	separate	groups	added	to	the	plankton	trophic	level.	The	diet	matrix	in	the	model	will	have	to	be	revised	accordingly	to	adequately	represent	new	plankton	groups	in	the	diets	of	all	consumers,	including	herring.	Predation on Pacific herring Predation	pressure	on	Pacific	herring	in	the	HG	model	was	distributed	over	all	the	stanzas	across	the	 four	 stocks	 in	 proportion	 to	 their	 biomasses	 in	 the	 revised-HG	 model.	 Thus,	 the	modifications	 to	 the	 diet	matrix	 and	 addition	 of	 new	 groups	 in	 the	 revised	 HG	model	 led	 to	changes	in	predators’	consumption	of	Pacific	herring	(Figure	15).	A Revised EwE Model of Haida Gwaii 		 53		Figure	15.	Consumption	of	Pacific	herring	by	all	its	predators	in	the	base	and	revised	HG	models.		New groups to be introduced to the HG model Saury The	Pacific	saury	(Cololabis	saira)	is	a	large	pelagic	forage	fish.	In	summer,	it	is	found	in	offshore	waters	as	far	north	as	the	Gulf	of	Alaska,	particularly	in	warm	years.	It	feeds	on	copepods	and	other	 zooplankton,	 and	 in	 turn	 is	prey	 to	many	 fishes,	 cetaceans	and	 seabirds.	 In	 the	 coming	years,	 saury	may	 become	 an	 increasingly	 important	 forage	 species	 in	 the	model	 area	 due	 to	increasing	sea	surface	temperatures.	2016  Fisheries Centre Research Reports 24(2) 		54	The	P/B,	Q/B	and	EE	parameter	values	and	diet	composition	for	this	group	were	obtained	from	data	 collected	 for	 an	Ecopath	model	of	 the	Alaska	Gyre	by	 Livingston	 (1996).	 The	biomass	of	saury	in	our	model	was	estimated	by	Ecopath	based	on	the	known	parameter	values. Epifaunal	invertebrates	This	 group	 in	 the	original	HG	model	was	 split	 into	 sea	urchins,	 other	 grazers,	 epifaunal	 filter-feeders,	 octopus	 and	 epifaunal	 carnivores	 to	 reflect	 the	 ecological	 diversity	 and	 internal	dynamics	of	the	original	group. Sea urchins These	 spiny,	 voracious	 echinoderms	 (class	 Echinoidea)	 are	 often	 the	 main	 herbivores	 in	intertidal	 and	 subtidal	 ecosystems,	 and	 can	 reduce	 kelp	 forests	 to	 so-called	 “urchin	 barrens”	unless	subject	to	top-down	control	by	abundant	sea	otters	(Estes	&	Palmisano	1974).	The	P/B,	Q/B	and	EE	parameter	values	and	diet	composition	for	this	group	were	obtained	from	an	Ecopath	model	of	Puget	 Sound	by	Harvey	et	 al.	 (2012).	 The	biomass	of	 sea	urchins	 in	our	model	was	 estimated	by	 Ecopath	based	on	 the	 known	parameter	 values. Moreover, we have allocated most of the landings from “other invert” fleet to this group, as indicated in Ainsworth (2006). Other grazers This	 group	 includes	 all	 herbivorous	 benthic	 invertebrates	 other	 than	 sea	 urchins,	 mainly	molluscs	(gastropods,	chitons)	and	small	crustaceans	(isopods,	amphipods	etc.).	The	P/B,	Q/B	and	EE	parameter	values	and	diet	composition	for	this	group	were	obtained	from	an	Ecopath	model	of	Puget	Sound	by	Harvey	et	al.	(2012).	The	biomass	of	other	grazers	in	our	model	was	estimated	by	Ecopath	based	on	the	known	parameter	values. Epifaunal filter-feeders This	group	comprises	those	sedentary	or	sessile	 invertebrates	 (mainly	bivalves,	barnacles,	and	tunicates	but	also	bryozoans,	brachiopods,	crinoids,	sabellid	polychaetes	etc.)	that	have	evolved	specialized	structures	to	filter	phytoplankton	and	detritus	from	the	water	column.	The	P/B,	Q/B	and	EE	parameter	values	and	diet	composition	for	this	group	were	obtained	from	biomass-weighted	averages	of	the	values	for	the	mussel,	geoduck,	barnacle	and	tunicate	groups	in	 an	Ecopath	model	of	 Puget	 Sound	by	Harvey	et	 al.	 (2012).	 The	biomass	of	 epifaunal	 filter-feeders	in	our	model	was	estimated	by	Ecopath	based	on	the	known	parameter	values.	Octopus This	 group	 includes	 the	 East	 Pacific	 red	 octopus	 (Octopus	 rubescens)	 and	 the	 giant	 Pacific	octopus	 (Enteroctopus	 dofleini).	 Both	 species	 are	 demersal	 predators	 feeding	 on	 large	crustaceans,	molluscs	and	small	fish,	with	the	latter	typically	hunting	larger	prey.	The	P/B,	Q/B	and	EE	parameter	values	and	diet	composition	for	this	group	were	obtained	from	an	Ecopath	model	of	Puget	Sound	by	Harvey	et	al.	(2012).	The	biomass	of	octopus	in	our	model	was	estimated	by	Ecopath	based	on	the	known	parameter	values. A Revised EwE Model of Haida Gwaii 		 55	Epifaunal carnivores This	group	includes	carnivorous	benthic	invertebrates	other	than	octopus	and	crabs	(mainly	sea	stars	and	predatory	gastropods).	The	P/B,	Q/B	and	EE	parameter	values	and	diet	composition	for	this	group	were	obtained	from	biomass-weighted	averages	of	the	values	for	the	sea	star	and	predatory	gastropod	groups	in	an	Ecopath	model	of	Puget	Sound	by	Harvey	et	al.	(2012).	The	biomass	of	epifaunal	carnivores	in	our	model	was	estimated	by	Ecopath	based	on	the	known	parameter	values. Macrozooplankton This	 group	 is	 composed	 of	 large	 non-gelatinous	 zooplankton	 excluding	 euphausiids	 and	amphipods	 (mainly	 pelagic	 shrimp,	 mysids	 and	 chaetognaths).	 It	 includes	 carnivorous	 and	omnivorous	species.	The	P/B,	Q/B	and	EE	parameter	values	and	diet	composition	for	this	group	were	obtained	from	an	Ecopath	model	of	Puget	Sound	by	Harvey	et	al.	(2012).	The	biomass	of	macrozooplankton	in	our	model	was	estimated	by	Ecopath	based	on	the	known	parameter	values. Pelagic amphipods This	group	includes	both	herbivorous	and	carnivorous	planktonic	amphipods.	The	P/B,	Q/B	and	EE	parameter	values	and	diet	composition	for	this	group	were	obtained	from	the	macrozooplankton	group	in	an	Ecopath	model	of	Puget	Sound	by	Harvey	et	al.	(2012).	The	biomass	 of	 pelagic	 amphipods	 in	 our	model	 was	 estimated	 by	 Ecopath	 based	 on	 the	 known	parameter	values. Small gelatinous zooplankton This	groups	comprises	small	zooplankton	whose	bodies	have	a	high	water	content	and	often	no	rigid	exoskeleton	(mainly	pteropods,	pelagic	tunicates,	ctenophores,	small	hydromedusae	etc.).	It	includes	carnivorous,	herbivorous	and	omnivorous	species.	The	P/B,	Q/B	and	EE	parameter	values	and	diet	composition	for	this	group	were	obtained	from	an	 Ecopath	 model	 of	 Puget	 Sound	 by	 Harvey	 et	 al.	 (2012).	 The	 biomass	 of	 small	 gelatinous	zooplankton	in	our	model	was	estimated	by	Ecopath	based	on	the	known	parameter	values.	Microzooplankton This	group	includes	all	heterotrophic	protists	(e.g.	ciliates,	some	dinoflagellates,	foraminiferans,	radiolarians	etc.)	as	well	as	 rotifers	and	other	microscopic	animals.	All	members	of	 this	group	feed	on	phytoplankton	and	detritus.	The	P/B,	Q/B	and	EE	parameter	values	and	diet	composition	for	this	group	were	obtained	from	an	Ecopath	model	of	Puget	Sound	by	Harvey	et	al.	(2012).	The	biomass	of	microzooplankton	in	our	model	was	estimated	by	Ecopath	based	on	the	known	parameter	values.	Eelgrass This	 group	 is	 comprised	 of	 the	 marine	 angiosperm	 Zostera	 marina,	 commonly	 known	 as	eelgrass.	 An	 ecosystem	 engineer	 in	 soft-bottom	 benthic	 communities,	 this	 species	 stabilizes	seafloor	sediments	and	provides	food	and	shelter	to	many	animals.	2016  Fisheries Centre Research Reports 24(2) 		56	The	P/B	and	EE	parameter	values	for	this	group	were	obtained	from	an	Ecopath	model	of	Puget	Sound	by	Harvey	et	al.	(2012).	The	biomass	of	eelgrass	in	our	model	was	estimated	by	Ecopath	based	on	the	known	parameter	values.	Macroalgae	This	group	in	the	original	HG	model	was	split	into	canopy	kelp	and	benthic	macroalgae	to	reflect	the	ecological	diversity	and	internal	dynamics	of	the	original	group. Canopy kelp This	 group	 includes	 the	 canopy-forming	 giant	 kelp	 (Macrocystis	 pyrifera)	 and	 bull	 kelp	(Nereocystis	luetkeana),	ecosystem	engineers	that	define	North	Pacific	kelp	forests.	The	P/B	and	EE	parameter	values	for	this	group	were	obtained	from	the	bull	kelp	group	 in	an	Ecopath	model	of	Puget	Sound	by	Harvey	et	al.	(2012).	The	biomass	of	canopy	kelp	in	our	model	was	estimated	by	Ecopath	based	on	the	known	parameter	values. Benthic macroalgae This	group	comprises	all	macroalgae	other	than	the	canopy-forming	kelps,	including	brown,	red	and	 green	 algae	 (Phaeophyceae,	 Rhodophyta	 and	 Chlorophyta).	 These	 species	 grow	 in	 the	understory	of	kelp	forests	and	in	other	rocky	subtidal	and	intertidal	habitats.	The	P/B	and	EE	parameter	values	for	this	group	were	obtained	from	an	Ecopath	model	of	Puget	Sound	by	Harvey	et	al.	(2012).	The	biomass	of	benthic	macroalgae	in	our	model	was	estimated	by	Ecopath	based	on	the	known	parameter	values. Benthic microalgae This	group	is	composed	of	benthic	diatoms	that	form	films	on	hard	substrates	in	both	intertidal	and	subtidal	zones	and	are	exploited	by	many	grazing	invertebrates.	The	P/B	and	EE	parameter	values	for	this	group	were	obtained	from	an	Ecopath	model	of	Puget	Sound	by	Harvey	et	al.	(2012).	The	biomass	of	benthic	microalgae	in	our	model	was	estimated	by	Ecopath	based	on	the	known	parameter	values.	Modifications to existing functional groups Humpback whales We	 examined	 two	 possible	 changes	 to	 this	 group’s	 diet	 composition	 in	 the	 Ecopath	 model,	based	on	the	trophic	levels	reported	for	these	whales	in	the	model	area	(3.5	for	NBC	and	3.4	for	SE	AK,	 no	 significant	 difference	 between	 regions)	 by	Witteveen	 et	 al.	 (2011)	 based	 on	 stable	isotope	 data.	 These	 values	 are	 substantially	 lower	 than	 that	 (3.65)	 derived	 from	 our	 original	Ecopath	model.	Ecopath	estimates	each	group’s	trophic	 level	based	on	 input	diet	composition	data,	 so	 a	 back-calculation	 can	 be	 used	 to	 estimate	 proportions	 of	 prey	 at	 different	 trophic	levels	 in	 predator	 diet.	 The	 analysis	 was	 based	 on	 two	 scenarios:	 (1)	 in	 which	 decreased	humpback	 whale	 trophic	 level	 (3.4)	 was	 due	 in	 equal	 measure	 to	 reduced	 consumption	 of	herring	and	other	forage	fish	and	(2)	in	which	the	same	trophic	level	was	obtained	by	reduced	consumption	of	 forage	fish	only,	 relative	to	the	diet	composition	 in	the	original	model	 (Figure	16).	Scenario	1	will	most	likely	be	used	in	the	final	version	of	the	model.	A Revised EwE Model of Haida Gwaii 		 57		Figure	16.	Proportions	of	herring,	forage	fish	and	euphausiids	in	humpback	whale	diets	and	the	total	ecosystem	biomass.	Minke whales The	biomass	density	of	this	group	was	slightly	decreased	for	the	sake	of	balancing	the	model,	but	remains	well	within	the	confidence	interval	of	the	local	abundance	estimates	published	by	Williams	 and	 Thomas	 (2007).	 As	 in	 the	 case	 of	 humpback	whales,	 the	 relative	 importance	 of	herring	 vs.	 other	 forage	 fish	 in	 the	 diet	 is	 somewhat	 conjectural	 and	 based	 on	 qualitative	knowledge	of	minke	whale	feeding	habits	in	British	Columbia	(Ford	2014).	0.000!0.100!0.200!0.300!0.400!0.500!0.600!0.700!Herring! Forage fish! Euphausiids!Proportion!Original model!Low TL, scenario 1!Low TL, scenario 2!Ecosystem!2016  Fisheries Centre Research Reports 24(2) 		58	References Ahrens,	R.	N.,	Walters,	C.	J.,	&	Christensen,	V.	(2012).	Foraging	arena	theory.	Fish	and	Fisheries,	13(1),	41-59.		Ainsworth,	C.	H.	(2006).	Strategic	marine	ecosystem	restoration	in	northern	British	Columbia	 (Doctor	of	Philosophy),	 Resource	 Management	 and	 Environmental	 Studies,	 The	 University	 of	 British	Columbia.				Ainsworth,	C.	H.,	Heymans,	 J.	 J.,	Pitcher,	T.,	&	Vasconcellos,	M.	 (2002).	Ecosystem	Models	of	Northern	British	 Columbia	 for	 the	 Time	 Periods	 2000,	 1950,	 1900	 and	 1750.	 Fisheries	 Centre	 Research	Reports,	Vol.	10	(4)	pp.	41.		Ainsworth,	 C.	 H.,	 Pitcher,	 T.	 J.,	 Heymans,	 J.	 J.,	 &	 Vasconcellos,	 M.	 (2008a).	 Reconstructing	 historical	marine	 ecosystems	 using	 food	 web	 models:	 northern	 British	 Columbia	 from	 pre-European	contact	to	present.	Ecological	Modelling,	216(3),	354-368.		Ainsworth,	 C.	H.,	 Varkey,	D.	 A.,	&	 Pitcher,	 T.	 J.	 (2008b).	 Ecosystem	 simulations	 supporting	 ecosystem-based	 fisheries	 management	 in	 the	 Coral	 Triangle,	 Indonesia.	 Ecological	 Modelling,	 214(2-4),	361-374.	doi:10.1016/j.ecolmodel.2008.02.039	Aráujo,	J.	N.,	&	Bundy,	A.	(2011).	Description	of	three	Ecopath	with	Ecosim	ecosystem	models	developed	for	 the	Bay	of	 Fundy,	Western	 Scotian	 Shelf	 and	NAFO	Division	4X.	DFO,	Can.	 Tech.	Rep.	 Fish.	Aquat.	Sci.	(2952:	xii	+	).		189	pp.	Barrie,	 J.	V.,	Conway,	K.	W.,	 Josenhans,	H.,	Clague,	 J.	 J.,	Mathewes,	R.	W.,	&	Fedje,	D.	W.	 (2005).	 Late	Quaternary	geology	of	Haida	Gwaii	 and	 surrounding	marine	areas.	 In	D.	 Fedje	&	R.	Mathewes	(Eds.),	Haida	Gwaii:	Human	History	and	Environment	 from	the	Time	of	Loon	to	the	Time	of	 the	Iron	People,	edited	by	DW	Fedje	and	RW	Mathewes	(pp.	7-20):	The	University	of	British	Columbia	Press,	Vancouver,	BC.	Beamish,	R.	J.,	&	McFarlane,	G.	A.	(1985).	Pacific	whiting,	Merluccius	productus,	stocks	off	the	west	coast	of	Vancouver	Island,	Canada.	Mar.	Fish.	Rev,	47(2),	75-81.		Beattie,	A.	(1999).	The	Hecate	Strait:	a	preliminary	model.	In	N.	Haggan	&	A.	Beattie	(Eds.):	Back	to	the	Future:	reconstructing	the	Hecate	Strait	ecosystem,	Fisheries	Center	Research	Reports,	Vol.	7	(3).		Beattie,	A.	(2001).	A	New	Model	for	Evaluating	the	Optimal	Size,	Placement	and	Configuration	of	Marine	Protected	Areas	 (Master	of	Science),	Department	of	Resource	Management	and	Environmental	Studies,	University	of	British	Columbia.				Blood,	D.	 A.	 (1993).	 Sea	Otter.	Wildlife	 Branch,	 BC	Ministry	 of	 Environment,	 Lands	 and	 Parks,	 Victoria	(ENV322358.1093).	Botsford,	 L.	 W.,	 Castilla,	 J.	 C.,	 &	 Peterson,	 C.	 H.	 (1997).	 The	 management	 of	 fisheries	 and	 marine	ecosystems.	Science,	277(5325),	509-515.		Calambokidis,	 J.,	 Barlow,	 J.,	 Ford,	 J.	 K.,	 Chandler,	 T.	 E.,	 &	 Douglas,	 A.	 B.	 (2009).	 Insights	 into	 the	population	 structure	 of	 blue	 whales	 in	 the	 eastern	 North	 Pacific	 from	 recent	 sightings	 and	photographic	identification.	Marine	Mammal	Science,	25(4),	816-832.		Cambria	 Gordon	 Ltd.	 (2006).	 Eulachon	 of	 the	 Pacific	 Northwest:	 A	 Life	 History.	 Living	 Landscapes	Program,	Royal	BC	Museum,	Victoria,	BC.	Cass,	A.	J.,	Beamish,	R.	J.,	&	McFarlane,	G.	A.	(1990).	Lingcod	(Ophiodon	elongatus).	Fisheries	and	Oceans,	Canada,	Can.	Spec.	Publ.	Fish.	Aquat.	Sci.	(109).		40	pp.	Chandler,	 P.	 C.,	 King,	 S.	 A.,	&	 Perry,	 R.	 I.	 (2015).	 State	 of	 the	 Physical,	 Biological	 and	 Selected	 Fishery	Resources	of	Pacific	Canadian	Marine	Ecosystems	in	2014.	Canadian	Technical	Report	of	Fisheries	and	Aquatic	Sciences	(3131).		vi+211	pp.	Christensen,	 T.,	 &	 Stafford,	 J.	 (2005).	 Raised	 beach	 archaeology	 in	 northern	 Haida	 Gwaii:	 preliminary	results	from	the	Cohoe	Creek	site.	In	D.	W.	Fedje	&	R.	W.	Mathewes	(Eds.),	Haida	Gwaii:	human	history	and	environments	from	the	time	of	the	loon	to	the	time	of	the	iron	people.	(pp.	204-244):	The	University	of	British	Columbia	Press,	Vancouver,	BC.	A Revised EwE Model of Haida Gwaii 		 59	Christensen,	V.	(1996).	Balancing	the	Alaska	Gyre	model.	In	D.	Pauly,	V.	Christensen,	&	N.	Haggan	(Eds.):	Mass-Balance	Models	 of	 North-eastern	 Pacific	 Ecosystems,	 Fisheries	 Centre	 Research	 Reports,	Vol.	4	(1)	pp.	32-35.		Christensen,	V.,	Coll,	M.,	Steenbeek,	J.,	Buszowski,	J.,	Chagaris,	D.,	&	Walters,	C.	J.	(2014).	Representing	variable	habitat	quality	in	a	spatial	food	web	model.	Ecosystems,	17(8),	1397-1412.		Christensen,	V.,	&	Pauly,	D.	(1992).	ECOPATH	II—a	software	for	balancing	steady-state	ecosystem	models	and	 calculating	 network	 characteristics.	 Ecological	 Modelling,	 61(3-4),	 169-185.	doi:10.1016/0304-3800(92)90016-8	Christensen,	 V.,	 &	 Walters,	 C.	 J.	 (2004).	 Ecopath	 with	 Ecosim:	 methods,	 capabilities	 and	 limitations.	Ecological	Modelling,	172(2-4),	109-139.		Christensen,	V.,	Walters,	C.	J.,	Pauly,	D.,	&	Forrest,	R.	(2008).	Ecopath	with	Ecosim	version	6	user	guide.	Lenfest	Ocean	Futures	Project,	University	of	British	Columbia,	Vancouver.		Common	 Jellyfish	 (Aurelia	 aurita).	 (n.d).	 Department	 of	 Fisheries	 and	 Aquaculture,	 Government	 of	Newfoundland	 and	 Labrador.	 Retrieved	 from	http://www.fishaq.gov.nl.ca/research_development/fdp/jellyfish.pdf	COSEWIC.	 (2002).	 COSEWIC	 assessment	 and	 update	 status	 report	 on	 the	 Blue	 Whale	 Balaenoptera	musculus	in	Canada.	Committee	on	the	Status	of	Endangered	Wildlife	in	Canada.	Ottawa.		vi	+	32	pp.	COSEWIC.	 (2003).	 COSEWIC	 assessment	 and	 status	 report	 on	 the	 sei	 whale	 Balaenoptera	 borealis	 in	Canada.	Ottawa:	Committee	on	the	Status	of	Endangered	Wildlife	in	Canada		vii	+	27	pp.	COSEWIC.	 (2005).	 COSEWIC	 assessment	 and	 update	 status	 report	 on	 the	 fin	 whale	 Balaenoptera	physalus	 in	Canada.	Ottawa:	Committee	on	the	Status	of	Endangered	Wildlife	 in	Canada	 	 ix+37	pp.	COSEWIC.	 (2009).	 COSEWIC	 assessment	 and	 status	 report	 on	 the	 Darkblotched	 Rockfish	 Sebastes	crameri	in	Canada.	Ottawa:	Committee	on	the	Status	of	Endangered	Wildlife	in	Canada		vii	+	48	pp.	Retrieved	from	www.sararegistry.gc.ca/status/status_e.cfm.	Cury,	 P.	M.,	 Shin,	 Y.-J.,	 Planque,	 B.,	 Durant,	 J.	M.,	 Fromentin,	 J.-M.,	 Kramer-Schadt,	 S.,	 .	 .	 .	 Grimm,	 V.	(2008).	 Ecosystem	oceanography	 for	 global	 change	 in	 fisheries.	Trends	 in	 Ecology	&	 Evolution,	23(6),	338-346.		Davis,	N.	D.,	Myers,	K.	W.,	&	Fournier,	W.	J.	(2009).	Winter	food	habits	of	Chinook	salmon	in	the	eastern	Bering	Sea.	N.	Pac.	Anadr.	Fish	Comm.	Bull,	5,	243-253.		DFO.	 (2013-02-13).	 Euphausiid.	 	 	 Retrieved	 August	 2015,	 from	 http://www.pac.dfo-mpo.gc.ca/science/species-especes/shellfish-coquillages/krill/index-eng.html	DFO.	(1998).	Dover	sole:	west	coast	Vancouver	Island	(areas	3C,	D)	to	Queen	Charlotte	Islands	(areas	5A-E).	DFO,	Science	Stock	Status	Report	(A6-04(1998)).	DFO.	(1999a).	English	sole	Hecate	Strait	(areas	DC/5),.	DFO,	S.	S.	S.	Report	(A6-05).	DFO.	(1999b).	Neon	flying	squid.	DFO,	Science	Stock	Status	Report	(C6-12	(1999)).	DFO.	(1999c).	Opal	squid.	DFO,	Science	Stock	Status	Report	(C6-04	(1999)).	DFO.	 (1999d).	 Pacific	 Ocean	 Perch	 British	 Columbia	 coast.	 DFO,	 Science	 Stock	 Status	 Report	 (A6-11(1999)).	DFO.	(2000a).	Dungeness	Crab,	Coastal	Fisheries,	Areas	B,	E,	G,	H,	I,	&	J.	DFO,	Science	Stock	Status	Report	(C6-14	(2000)).	DFO.	(2000b).	Inshore	Rockfish.	DFO,	Science	Stock	Status	Report	(A6-15(2000)).	DFO.	(2009a).	Canadian	assessment	of	Pacific	hake	in	U.S.	and	Canadian	waters	 in	2009.	DFO,	Can.	Sci.	Advis.	Sec.	Sci.	Resp.	(2009/002).	DFO.	 (2009b).	 Recovery	 potential	 assessment	 of	 humpback	 whales,	 Pacific	 population.	 DFO,	 Can.	 Sci.	Advis.	Sec.	Sci.	Advis.	Rep.	(2009/048).	2016  Fisheries Centre Research Reports 24(2) 		60	DFO.	 (2010a).	 Chinook	 salmon	 abundance	 levels	 and	 survival	 of	 resident	 killer	 whales.	 DFO,	 Can.	 Sci.	Advis.	Sec.	Sci.	Advis.	Rep.	(2009/075).	DFO.	(2010b).	Pacific	region	cold-water	coral	and	sponge	conservation	strategy	2010-2015.	Fisheries	and	Oceans	Canada	(DFO/2010-1663).	DFO.	 (2011a).	 Management	 procedures	 for	 the	 multi-gear	 sablefish	 (Anoplopoma	 fimbria)	 fishery	 in	British	Columbia,	Canada.	DFO,	Can.	Sci.	Advis.	Sec.	Sci.	Advis.	Rep.	(2011/025).	DFO.	 (2011b,	 Last	 updated:	 2011-04-26).	 Zooplankton	 Laboratory.	 	 	 Retrieved	 Septermber	 2015,	 from	http://www.pac.dfo-mpo.gc.ca/science/oceans/plankton-plancton/zoolab-eng.html	DFO.	 (2012a).	 Information	 relevant	 to	 the	 assessment	 of	 critical	 habitat	 for	 Blue,	 Fin,	 Sei	 and	 North	Pacific	Right	Whales	in	British	Columbia.	DFO,	Can.	Sci.	Advis.	Sec.	Sci.	Advis.	Rep.	(2011/078).		12	pp.	DFO.	(2012b).	Lingcod	(Ophiodon	Elongatus)	stock	assessment	and	yield	advice	for	outside	stocks	in	British	Columbia.	DFO,	Can.	Sci.	Advis.	Sec.	Sci.	Advis.	Rep.	(2011/051).	DFO.	 (2013a).	 Management	 Areas-Pacific	 Region.	 	 	 Retrieved	 2013,	 from	 http://www.pac.dfo-mpo.gc.ca/fm-gp/maps-cartes/areas-secteurs/index-eng.html	DFO.	(2013b).	Pacific	Ocean	Perch	(Sebastes	alutus)	stock	assessments	for	the	west	coast	of	Vancouver	Island	and	the	north	and	west	coasts	of	Haida	Gwaii,	British	Columbia.	DFO,	Can.	Sci.	Advis.	Sec.	Sci.	Advis.	Rep.	(2013/038).	DFO.	 (2014,	 2014-11-18).	 Statistics:	 commercial	 fisheries	 landings.	 	 	 Retrieved	 20145,	 from	http://www.dfo-mpo.gc.ca/stats/commercial/sea-maritimes-eng.htm	DFO.	 (2015a).	Assessment	of	Pacific	Cod	 (Gadus	Macrocephalus)	 for	Hecate	 Strait	 (Area	5CD)	 in	2013.	DFO,	Can.	Sci.	Advis.	Sec.	Sci.	Advis.	Rep.	(2015/015).	DFO.	 (2015b,	 Last	 updated:	 2015-12-09).	 Commercial	 fisheries.	 	 	 Retrieved	 Dec	 2015,	 from	http://www.dfo-mpo.gc.ca/stats/commercial/sea-maritimes-eng.htm	DFO.	 (2015c).	 Stock	 assessment	 and	management	 advice	 for	 BC	Pacific	 herring:	 2015	 status	 and	 2016	forecast.	DFO,	Can.	Sci.	Advis.	Sec.	Sci.	Resp.	(2015/038).	DFO.	 (2015d).	 Stock	 assessment	 and	 management	 advice	 for	 British	 Columbia	 Pacific	 Herring:	 2014	status	and	2015	forecast.	DFO,	Can.	Sci.	Advis.	Sec.	Sci.	Advis.	Rep.	(2014/060).	DFO.	 (2016).	 Integrated	 fisheries	 management	 plan	 summary:	 shrimp	 trawl,	 Smooth	 Pink	 Shrimp	(Pandalus	 jordani)	 and	 Sidestripe	 Shrimp	 (Pandalopsis	 dispar),Pacific	 Region.	 Retrieved	 from	http://www.pac.dfo-mpo.gc.ca/fm-gp/mplans/2016/shrimp-trawl-crevette-chalut-sm-2016-17-eng.pdf	DFO.	 (n.d).	 Herring	 Catch	 Tables.	 	 	 Retrieved	 Sept	 2015,	 from	 http://www.pac.dfo-mpo.gc.ca/science/species-especes/pelagic-pelagique/herring-hareng/herspawn/tabcfram-eng.html	DFO.	 (n.d.).	 2013-2017	 Euphausiids	 Integrated	 Fisheries	 Management	 Plan.	 	 Retrieved	 from	http://www.pac.dfo-mpo.gc.ca/fm-gp/mplans/2013/krill-2013-17-eng.pdf.	Duffy,	 J.	 (2015).	 Marine	 biodiversity	 and	 food	 security.	 	 	 Retrieved	 August	 2015,	 from	http://www.eoearth.org/view/article/154466	Estes,	J.	A.,	&	Palmisano,	J.	F.	(1974).	Sea	otters:	their	role	in	structuring	nearshore	communities.	Science,	185(4156),	1058-1060.		Fargo,	J.,	&	Starr,	P.	J.	(2001).	Turbot	stock	assessment	for	2001	and	recommendations	for	management	2002.	DFO,	Can.	Sci.	Advis.	Sec.	Res.	Doc.	(2001/150).	Fedje,	D.	W.,	Mackie,	A.	P.,	Wigen,	R.	J.,	Mackie,	Q.,	&	Lake,	C.	(2005a).	Kilgii	Gwaay:	An	early	maritime	site	 in	the	south	of	Haida	Gwaii	Haida	Gwaii:	human	history	and	environment	from	the	time	of	the	loon	to	the	time	of	the	iron	people.	 (pp.	187-203):	The	University	of	British	Columbia	Press,	Vancouver,	BC.	A Revised EwE Model of Haida Gwaii 		 61	Fedje,	D.	W.,	&	Mackie,	Q.	(2005).	Overview	of	cultural	history.	In	D.	Fedje	&	R.	Mathewes	(Eds.),	Haida	Gwaii:	human	history	and	environment	from	the	time	of	the	loon	to	the	time	of	the	iron	people.	(pp.	154-162):	The	University	of	British	Columbia	Press,	Vancouver,	BC.	Fedje,	D.	W.,	Magne,	M.	P.,	&	Christensen,	T.	(2005b).	Test	Excavation	at	Raised	Beach	Sites	in	Southern	Haida	Gwaii.	 In,	DW	Fedje	and	RW	Mathewes.	 In	D.	Fedje	&	R.	Mathewes	 (Eds.),	Haida	Gwaii:	human	history	and	environment	 from	 the	 time	of	 the	 loon	 to	 the	 time	of	 the	 iron	people.	 (pp.	204-244):	The	University	of	British	Columbia	Press,	Vancouver,	BC.	Flinn,	R.	D.,	Trites,	A.	W.,	Gregr,	E.	J.,	&	Perry,	R.	 I.	(2002).	Diets	of	fin,	sei,	and	sperm	whales	in	British	Columbia:	 An	 analysis	 of	 commercial	 whaling	 records,	 1963-1967.	Marine	 Mammal	 Science,	18(3),	663-679.		Ford,	 J.	 K.,	 Ellis,	 G.	M.,	 Barrett-Lennard,	 L.	 G.,	Morton,	 A.	 B.,	 Palm,	 R.	 S.,	 &	 Balcomb	 III,	 K.	 C.	 (1998).	Dietary	 specialization	 in	 two	 sympatric	 populations	 of	 killer	 whales	 (Orcinus	 orca)	 in	 coastal	British	Columbia	and	adjacent	waters.	Canadian	Journal	of	Zoology,	76(8),	1456-1471.		Ford,	 J.	 K.	 B.	 (2014).	 Marine	 Mammals	 of	 British	 Columbia.	 Royal	 British	 Columbia	 Museum	 Press,	Victoria,	BC.		Ford,	J.	K.	B.,	Rambeau,	A.	L.,	Abernethy,	R.	M.,	Boogaards,	M.	D.,	Nichol,	L.	M.,	&	Spaven,	L.	D.	(2009).	An	assessment	of	the	potential	for	recovery	of	humpback	whales	off	the	Pacific	coast	of	Canada.	DFO,	Canadian	Science	Advisory	Secretariat	Research	Report	(2009/015).		iv+33	pp.	Fouzai,	N.,	Coll,	M.,	Palomera,	I.,	Santojanni,	A.,	Arneri,	E.,	&	Christensen,	V.	(2012).	Fishing	management	scenarios	 to	 rebuild	 exploited	 resources	 and	 ecosystems	 of	 the	 Northern-Central	 Adriatic	(Mediterranean	Sea).	Journal	of	Marine	Systems,	102,	39-51.		Fredin,	R.	A.	 (1985).	Pacific	 cod	 in	 the	eastern	Bering	 sea:	a	 sysnopsis.	Northwest	and	Alaska	Fisheries	Center,	 National	 Marine	 Fisheries	 Service,	 U.S	 Department	 of	 Commerce	 (NWAFC	 processed	report	85-05).	Gregr,	 E.	 J.	 (2004).	Marine	mammals	 in	 the	Hecate	 Strait	 ecosystem.	Can.	 Tech.	 Rep.	 Fish.	 Aquat.	 Sci.		2503	:2556	pp.	Gregr,	E.	J.,	Nichol,	L.,	Ford,	J.	K.,	Ellis,	G.,	&	Trites,	A.	W.	(2000).	Migration	and	population	structure	of	northeastern	 Pacific	 whales	 off	 coastal	 British	 Columbia:	 An	 analysis	 of	 commercial	 whaling	records	from	1908-1967.	Marine	Mammal	Science,	16(4),	699-727.		Gregr,	E.	J.,	&	Trites,	A.	W.	(2001).	Predictions	of	critical	habitat	for	five	whale	species	 in	the	waters	of	coastal	British	Columbia.	Canadian	Journal	of	Fisheries	and	Aquatic	Sciences,	58(7),	1265-1285.		Guénette,	 S.	 (2005).	 Models	 of	 Southeast	 Alaska.	 In	 S.	 Guénette	 &	 V.	 Christensen	 (Eds.):	 Food	 web	models	and	data	for	studying	fisheries	and	environmental	impacts	on	Eastern	Pacific	ecosystems,	Fisheries	Centre	Research	Reports,	Vol.	13	(1)	pp.	106-178.		Harvey,	 C.	 J.,	Williams,	G.	 D.,	&	 Levin,	 P.	 S.	 (2012).	 Food	web	 structure	 and	 trophic	 control	 in	 central	Puget	Sound.	Estuaries	and	Coasts,	35(3),	821-838.		Hay,	 D.	 (1985).	 Reproductive	 biology	 of	 Pacific	 herring	 (Clupea	 harengus	 pallasi).	 Canadian	 Journal	 of	Fisheries	and	Aquatic	Sciences,	42(S1),	s111-s126.		Hay,	D.	E.,	&	McCarter,	P.	B.	(2013).	Herring	spawning	areas	of	British	Columbia:	A	review,	geographical	analysis	and	classification.	DFO,	(Revised	edition:	2013).	Heise,	K.,	Sloan,	N.,	Olesiuk,	P.,	Bartier,	P.,	&	Ford,	 J.	 (2003).	Living	Marine	Legacy	of	Gwaii	Haanas.	 IV:	Marine	Mammal	Baseline	to	2003	and	Marine	Mammal-related	Management	Issues	throughout	the	 Haida	 Gwaii	 Region.	 Parks	 Canada	 Technical	 Reports	 in	 Ecosystem	 Science	 (No.	 38:	 xii	 +	150pp).		150	pp.	Helser,	T.	E.,	Stewart,	I.	J.,	&	Hamel,	O.	S.	(2008).	Stock	Assessment	of	Pacific	Hake	(Whiting)	in	U.S.	and	Canadian	Waters	in	2008	Northwest	Fisheries	Science	Center,	National	Marine	Fisheries	Service,	National	Oceanic	and	Atmospheric	Administration,	Seattle,	Washington	(2008).	2016  Fisheries Centre Research Reports 24(2) 		62	Hulbert,	L.,	Aires-da-Silva,	A.,	Gallucci,	V.,	&	Rice,	J.	(2005).	Seasonal	foraging	movements	and	migratory	patterns	 of	 female	 Lamna	 ditropis	 tagged	 in	 Prince	 William	 Sound,	 Alaska.	 Journal	 of	 Fish	Biology,	67(2),	490-509.		IPHC.	 (1998).	 The	 Pacific	 halibut:	 biology,	 fishery,	 and	 management.	 Internation	 Pacific	 Halibut	Commission,	Seattle,	Washington	98145-2009,	USA	(Technical	Report	No	40,	ISSN:	0579-3920).	Kaiser,	 G.	 W.	 (2002).	 Seabirds	 in	 Northern	 British	 Columbia.	 In	 T.	 Pitcher,	 J.	 J.	 Heymans,	 &	 M.	Vasconcellos	 (Eds.):	 Information	 Supporting	 Past	 and	 Present	 Ecosystem	 Models	 of	 Northern	British	Columbia	and	the	Newfoundland	Shelf,	Fisheries	Centre	Research	Reports,	Vol.	10	(1)	pp.	85-87.		Livingston,	P.	(1996).	Alaska	Gyre	model:	Miscellaneous	fishes.	In	D.	Pauly	&	V.	Christensen	(Eds.):	Mass-balance	models	of	Northeastern	Pacific	ecosystems,	Fisheries	Centre	Research	Reports,	Vol.	4	(1)	pp.	25.		Lucas,	B.	(2011).	Are	jellyfish	the	food	of	the	future?	INFOFISH	International,	4,	60-63.		Martell,	S.	J.,	Essington,	T.	E.,	Lessard,	B.,	Kitchell,	J.	F.,	Walters,	C.	J.,	&	Boggs,	C.	H.	(2005).	Interactions	of	productivity,	predation	risk,	and	fishing	effort	in	the	efficacy	of	marine	protected	areas	for	the	central	Pacific.	Canadian	Journal	of	Fisheries	and	Aquatic	Sciences,	62(6),	1320-1336.		Matkin,	D.	R.,	 Straley,	 J.	M.,	&	Gabriele,	C.	M.	 (2007).	 Killer	whale	 feeding	ecology	and	non-predatory	interactions	with	other	marine	mammals	in	the	Glacier	Bay	region	of	Alaska.	In	J.	F.	Piatt	&	S.	M.	Gende	 (Eds.):	Proceedings	of	 the	Fourth	Glacier	Bay	Science	Symposium,	October	26–28,	2004	U.S.	Geological	Survey	Scientific	Investigations	Report	2007-5047,	pp.	155-158.		McFarlane,	G.,	&	Beamish,	R.	(1999).	Sardines	return	to	British	Columbia	waters.	PICES	scientific	report.	Sidney	BC(10),	77-82.		McFarlane,	G.	A.,	King,	J.	R.,	&	Beamish,	R.	J.	(2000).	Have	there	been	recent	changes	in	climate?	Ask	the	fish.	Progress	in	Oceanography,	47(2),	147-169.		Mizroch,	 S.	 A.,	 Rice,	 D.	 W.,	 Zwiefelhofer,	 D.,	 Waite,	 J.,	 &	 Perryman,	 W.	 L.	 (2009).	 Distribution	 and	movements	of	fin	whales	in	the	North	Pacific	Ocean.	Mammal	Review,	39(3),	193-227.		Moody,	M.F.	and	Pitcher	T.J.	(2010)	Eulachon	(Thaleichthys	pacificus):	past	and	present.	Fisheries	Centre	Research	Reports	18(2):	197	pp.	Morissette,	L.,	Kaschner,	K.,	&	Gerber,	L.	R.	(2010).	Ecosystem	models	clarify	the	trophic	role	of	whales	off	Northwest	Africa.	Marine	Ecology	Progress	Series,	404,	289-302.		Nakano,	H.,	&	Seki,	M.	 (2003).	Synopsis	of	biological	data	on	the	blue	shark,	Prionace	glauca	Linnaeus.	Bulletin	of	Fisheries	Research	Agency	(Japan),	6,	18-55.		Newlands,	N.	(1998).	Box	3:	Salmon	Population	Parameters.	In	D.	Pauly,	T.	Pitcher,	&	D.	Preikshot	(Eds.):	Back	 to	 the	 Future:	 Reconstructing	 the	 Strait	 of	 Georgia	 Ecosystem	 Fisheries	 Centre	 Research	Reports,	Vol.	6	(5)	pp.	77.		Nichol,	L.	M.,	Abernethy,	R.,	Flostrand,	L.,	Lee,	T.	S.,	&	Ford,	J.	K.	B.	(2010).	Information	relevant	for	the	identification	of	Critical	Habitats	of	North	Pacific	Humpback	Whales	(Megaptera	novaeangliae)	in	British	Columbia.	DFO,	Can.	Sci.	Advis.	Sec.	Res.	Doc.	(2009/116:	iv	+).		40	pp.	NMFS.	(2011).	Final	Recovery	Plan	for	the	Sei	Whale	(Balaenoptera	borealis).	National	Marine	Fisheries	Service,	Office	of	Protected	Resources,	Silver	Spring,	MD.		108	pp.	NMFS.	 (2014).	 Status	 Review	 of	 Southeast	 Alaska	 Herring	 (Clupea	 pallasi),	 Threats	 Evaluation	 and	Extinction	Risk	Analysis.	National	Marine	Fisheries	Service,	Office	of	Protected	Resources		183	pp.	Okey,	T.	A.,	&	Pauly,	D.	(1998).	In	T.	A.	Okey	&	D.	Pauly	(Eds.):	Trophic	Mass-Balance	Model	of	Alaska’s	Prince	William	Sound	Ecosystem,	for	the	Post-Spill	Period	1994-1996,	Fisheries	Centre	Research	Reports,	Vol.	7	(4)	pp.	1-135.		Orchard,	T.	J.	(2011).	Late	Holocene	fisheries	in	Gwaii	Haanas:	Species	composition,	trends	in	abundance,	and	environmental	or	cultural	explanations.	In	M.	L.	Moss	&	A.	Cannon	(Eds.),	The	archaeology	of	North	Pacific	fisheries	(pp.	111-128):	Fairbanks,	AK:	University	of	Alaska	Press.	A Revised EwE Model of Haida Gwaii 		 63	Palsson,	W.	A.,	Hoffmann,	S.,	Clarke,	P.,	&	Beam,	J.	 (2003).	Results	 from	the	2001	transboundary	trawl	survey	of	the	southern	Strait	of	Georgia,	San	Juan	Archipelago	and	adjacent	waters.	Washington	Department	of	Fish	and	Wildlife,	Fish	Program,	Fish	Management	Division.	Pauly,	D.,	Christensen,	V.,	&	Walters,	C.	 (2000).	 Ecopath,	 Ecosim,	and	Ecospace	as	 tools	 for	evaluating	ecosystem	 impact	 of	 fisheries.	 ICES	 Journal	 of	 Marine	 Science,	 57(3),	 697-706.	doi:10.1006/jmsc.2000.0726	Pauly,	D.,	Trites,	A.,	Capuli,	E.,	&	Christensen,	V.	 (1998).	Diet	composition	and	trophic	 levels	of	marine	mammals.	ICES	Journal	of	Marine	Science:	Journal	du	Conseil,	55(3),	467-481.		Pitcher,	 T.	 J.	 (2001).	 Fisheries	managed	 to	 rebuild	ecosystems?	Reconstructing	 the	past	 to	 salvage	 the	future.	Ecological	Applications,	11(2),	601-617.		Pitcher,	 T.	 J.,	 Buchary,	 E.,	 &	 Trujillo,	 P.	 (2002)Spatial	 simulations	 of	 Hong	 Kong's	 marine	 ecosystem:	ecological	 and	 economic	 forecasting	 of	 marine	 protected	 areas	 with	 human-made	 reefs,	Fisheries	Centre	Research	Report,	Vol.	10	(3).	University	of	British	Columbia.		Polovina,	J.	J.	(1984).	Model	of	a	coral	reef	ecosystem	I.	the	ecopath	model	and	its	application	to	french	frigate	 shoals.	 Coral	 Reefs,	 3(1),	 1-11.	 	 Retrieved	 from	http://download.springer.com.ezproxy.library.ubc.ca/static/pdf/460/art%253A10.1007%252FBF00306135.pdf?auth66=1390535512_16e6b761a5ffddac157bc6b9c30298b0&ext=.pdf	Preikshot,	D.	(2005).	Data	sources	and	derivation	of	parameters	for	generalised	Northeast	Pacific	Ocean	Ecopath	with	Ecosim	models.	In	V.	Christensen	&	S.	Guénette	(Eds.):	Food	web	models	and	data	for	studying	fisheries	and	environmental	impacts	on	Eastern	Pacific	ecosystems,	Fisheries	Centre	Research	Reports,	Vol.	13	(1)	pp.	179-206.		Raum-Suryan,	 K.,	 Pitcher,	 K.,	 &	 Lamy,	 R.	 (2004).	 Sea	 Otter,	 Enhydra	 lutris,	 Sightings	 off	 Haida	Gwaii/Queen	 Charlotte	 Islands,	 British	 Columbia,	 1972-2002.	 The	 Canadian	 Field-Naturalist,	118(2),	270-272.		Reilly,	S.	B.	(1984).	Observed	and	maximum	rates	of	increase	in	gray	whales,	Eschrichtius	robustus.	Rep.	Int.	Whaling	Comm.	Spec.	Issue,	6,	389-400.		Richardson,	 A.	 J.,	 Bakun,	 A.,	 Hays,	 G.	 C.,	 &	 Gibbons,	 M.	 J.	 (2009).	 The	 jellyfish	 joyride:	 causes,	consequences	 and	 management	 responses	 to	 a	 more	 gelatinous	 future.	 Trends	 in	 Ecology	 &	Evolution,	24(6),	312-322.		Roberts,	J.	M.,	Wheeler,	A.	J.,	&	Freiwald,	A.	(2006).	Reefs	of	the	deep:	the	biology	and	geology	of	cold-water	coral	ecosystems.	Science,	312(5773),	543-547.		Salomon,	A.	K.,	Waller,	N.	P.,	McIlhagga,	C.,	Yung,	R.	L.,	&	Walters,	C.	(2002).	Modeling	the	trophic	effects	of	marine	protected	area	zoning	policies:	a	case	study.	Aquatic	Ecology,	36(1),	85-95.		Schweigert,	J.	F.,	Boldt,	J.	L.,	Flostrand,	L.,	&	Cleary,	J.	S.	(2010).	A	review	of	factors	limiting	recovery	of	Pacific	herring	stocks	in	Canada.	ICES	Journal	of	Marine	Science:	Journal	du	Conseil,	fsq134.		Sinclair,	 A.	 F.,	 Martell,	 S.,	 &	 Boutillier,	 J.	 (2001).	 Assessment	 of	 Pacific	 Cod	 off	 the	 West	 Coast	 of	Vancouver	 Island	 and	 in	 Hecate	 Strait,	 November	 2001.	 DFO,	 Canadian	 Science	 Advisory	Secretariat	Research	Document	(2001/159).	Sloan,	N.	A.,	&	Gunn,	C.	R.	(1985).	Fishing,	processing,	and	marketing	of	the	jellyfish,	Aurelia	aurita	(L.),	from	southern	British	Columbia.	DFO,	Canadian	Industry	Report	of	Fisheries	and	Aquatic	Sciences	No.	157.	Southern	BC	shelf	model.	(1996).	In	D.	Pauly	&	V.	Christensen	(Eds.):	Mass-Balance	Models	of	the	North-Eastern	Pacific,	Fisheries	Centre	Research	Reports,	Vol.	4	(1).		Springer,	 A.	M.	 (1992).	 A	 review:	Walleye	 pollock	 in	 the	 North	 Pacific–how	much	 difference	 do	 they	really	make?	Fisheries	Oceanography,	1(1),	80-96.	doi:10.1111/j.1365-2419.1992.tb00026.x	Steenbeek,	 J.,	 Coll,	 M.,	 Gurney,	 L.,	 Mélin,	 F.,	 Hoepffner,	 N.,	 Buszowski,	 J.,	 &	 Christensen,	 V.	 (2013).	Bridging	 the	 gap	 between	 ecosystem	 modeling	 tools	 and	 geographic	 information	 systems:	2016  Fisheries Centre Research Reports 24(2) 		64	Driving	a	 food	web	model	with	external	spatial–temporal	data.	Ecological	Modelling,	263,	139-151.		Stewart,	I.	J.,	&	Hamel,	O.	S.	(2010).	Stock	assessment	of	Pacific	hake,	Merluccius	productus,(aka	whiting)	in	 US	 and	 Canadian	 waters	 in	 2010	 	 DRAFT).	 Northwest	 Fisheries	 Science	 Center,	 National	Oceanic	 and	 Atmospheric	 Administration	 Seattle,	 WA	 98112,	 USA.	 Retrieved	 from	http://dev.pcouncil.org/wp-content/uploads/E3a_ATT2_HAKE_USCAN_NWFSC_MARCH_2010_BB.pdf	Surma,	S.,	&	Pitcher,	T.	J.	(2015).	Predicting	the	effects	of	whale	population	recovery	on	Northeast	Pacific	food	webs	and	fisheries:	an	ecosystem	modelling	approach.	Fisheries	Oceanography,	24(3),	291-305.		Szpak,	 P.,	 Orchard,	 T.	 J.,	 &	 Gröcke,	 D.	 R.	 (2009).	 A	 late	 holocene	 vertebrate	 food	web	 from	 southern	Haida	 Gwaii	 (Queen	 Charlotte	 Islands,	 British	 Columbia).	 Journal	 of	 Archaeological	 Science,	36(12),	2734-2741.		Tranquilla,	 L.	M.,	Truman,	K.,	 Johannessen,	D.,	&	Hooper,	T.	 (2007).	Appendix	K:	Marine	Birds.	 In	B.	G.	Lucas,	 S.	 Verrin,	 &	 R.	 Brown	 (Eds.):	 Ecosystem	 overview:	 Pacific	 North	 Coast	 Integrated	Management	Area	(PNCIMA),	Can.	Tech.	Rep.	Fish.	Aquat.	Sci.,	pp.	2667:	vi	+	2668.		Trites,	A.	W.,	&	Pauly,	D.	(1998).	Estimating	mean	body	masses	of	marine	mammals	from	maximum	body	lengths.	Canadian	Journal	of	Zoology,	76(5),	886-896.		Varkey,	 D.,	 Ainsworth,	 C.	 H.,	 &	 Pitcher,	 T.	 J.	 (2012).	 Modelling	 Reef	 Fish	 Population	 Responses	 to	Fisheries	Restrictions	in	Marine	Protected	Areas	in	the	Coral	Triangle.	Journal	of	Marine	Biology,	1-18.	doi:10.1155/2012/721483	Vasconcellos,	 M.,	 &	 Pitcher,	 T.	 (2002).	 Historic	 reference	 points	 for	 models	 of	 past	 ecosystems	 in	northern	 British	 Columbia.	 In	 T.	 Pitcher,	 J.	 J.	 Heymans,	 &	M.	 Vasconcellos	 (Eds.):	 Information	Supporting	 Past	 and	 Present	 Ecosystem	 Models	 of	 Northern	 British	 Columbia	 and	 the	Newfoundland	Shelf,	Fisheries	Centre	Research	Reports,	Vol.	10	(1)	pp.	116.		Wada,	Y.,	&	Kelson,	J.	(1996).	Seabirds	of	the	southern	B.C.	shelf.	 In	V.	Christensen	&	N.	Haggan	(Eds.):	Mass-Balance	Models	 of	 North-eastern	 Pacific	 Ecosystems,	 Fisheries	 Centre	 Research	 Reports,	Vol.	4	(1)	pp.	55-56.		Walters,	 C.,	 Christensen,	 V.,	 &	 Pauly,	 D.	 (1997).	 Structuring	 dynamic	models	 of	 exploited	 ecosystems	from	trophic	mass-balance	assessments.	Reviews	in	Fish	Biology	and	Fisheries,	7(2),	139-172.		Walters,	C.,	Christensen,	V.,	Walters,	W.,	&	Rose,	K.	(2010).	Representation	of	multistanza	life	histories	in	Ecospace	models	 for	 spatial	 organization	of	 ecosystem	 trophic	 interaction	patterns.	Bulletin	of	Marine	Science,	86(2),	439-459.		Walters,	 C.,	 Pauly,	 D.,	 &	 Christensen,	 V.	 (1999).	 Ecospace:	 prediction	 of	mesoscale	 spatial	 patterns	 in	trophic	relationships	of	exploited	ecosystems,	with	emphasis	on	the	impacts	of	marine	protected	areas.	Ecosystems,	2(6),	539-554.		Walters,	 C.,	 Pauly,	 D.,	 Christensen,	 V.,	 &	 Kitchell,	 J.	 F.	 (2000).	 Representing	 density	 dependent	consequences	of	life	history	strategies	in	aquatic	ecosystems:	EcoSim	II.	Ecosystems,	3(1),	70-83.		Retrieved	 from	http://download.springer.com.ezproxy.library.ubc.ca/static/pdf/874/art%253A10.1007%252Fs100210000011.pdf?auth66=1390536329_d8f48c7e7e7b10765d3c0ebcf433e443&ext=.pdf	Weng,	K.	C.,	Foley,	D.	G.,	Ganong,	J.	E.,	Perle,	C.,	Shillinger,	G.	L.,	&	Block,	B.	A.	(2008).	Migration	of	an	upper	 trophic	 level	 predator,	 the	 salmon	 shark	 Lamna	 ditropis,	 between	 distant	 ecoregions.	Marine	Ecology	Progress	Series,	372,	253-264.		Wespestad,	V.	G.	(1982).	Cohort	analysis	of	catch	data	on	Pacific	herring	in	the	eastern	Bering	Sea,	1959-81	 	 National	 Oceanic	 and	 Atmospheric	 Administration,Technical	 Memorandum	 (NOAA-TM-NMFS-F/NWC-24)).	Seattle.		A Revised EwE Model of Haida Gwaii 		 65	Wigen,	 R.	 J.	 (2005).	 History	 of	 the	 vertebrate	 fauna	 in	 Haida	Gwaii.	 In	 D.	 Fedje	&	 R.	Mathewe	 (Eds.),	Haida	Gwaii:	human	history	and	environment	 from	the	time	of	 the	 loon	to	the	time	of	 the	 iron	people.	(pp.	96-115):	The	University	of	British	Columbia	Press,	Vancouver,	BC.	Williams,	 R.,	 Okey,	 T.	 A.,	Wallace,	 S.,	 &	 Gallucci,	 V.	 F.	 (2010).	 Shark	 aggregation	 in	 coastal	 waters	 of	British	Columbia.	Marine	Ecology	Progress	Series,	414,	249-256.		Williams,	R.,	&	Thomas,	L.	(2007).	Distribution	and	abundance	of	marine	mammals	in	the	coastal	waters	of	British	Columbia,	Canada.	Journal	of	Cetacean	Research	and	Management,	9(1),	15-28.		Witteveen,	 B.	H.,	Worthy,	G.	A.,	Wynne,	 K.	M.,	Hirons,	A.	 C.,	 Andrews,	A.	G.,	&	Markel,	 R.	W.	 (2011).	Trophic	 levels	of	North	Pacific	humpback	whales	 (Megaptera	novaeangliae)	through	analysis	of	stable	isotopes:	implications	on	prey	and	resource	quality.	Aquatic	Mammals,	37(2),	101.		Zeller,	 D.,	 &	 Reinert,	 J.	 (2004).	 Modelling	 spatial	 closures	 and	 fishing	 effort	 restrictions	 in	 the	 Faroe	Islands	marine	ecosystem.	Ecological	Modelling,	172(2),	403-420.					 	2016  Fisheries Centre Research Reports 24(2) 		66	Appendices Appendix A. Revised HG ecosystem model: parameterization Shaded	values	were	estimated	by	Ecopath.	SN	 Group	name	 TL	 B	(t/km²)	 Z	(/year)	 P/B	(/year)	 Q/B	(/year)	 EE	 P/Q	(/year)	1	 Sea	Otters	 3.188	 0.000		0.130	 101.500	 0.000	 0.001	2	 Gray	whales	 3.025	 0.030		0.060	 5.300	 0.089	 0.011	3	 Humpback	whales	 3.587	 0.185		0.060	 4.600	 0.000	 0.013	4	 Minke	whales	 3.594	 0.030		0.090	 6.300	 0.059	 0.014	5	 Blue	whales	 3.325	 0.006		0.040	 3.500	 0.000	 0.011	6	 Fin	whales	 3.435	 0.035		0.050	 4.100	 0.000	 0.012	7	 Sei	whales	 3.373	 0.000		0.060	 5.200	 0.000	 0.012	8	 Sperm	whales	 4.091	 0.010		0.050	 5.100	 0.000	 0.010	9	 Resident	orcas	 4.739	 0.003		0.090	 7.700	 0.000	 0.012	10	 Transient	orcas	 5.137	 0.002		0.090	 7.700	 0.000	 0.012	11	 Small	odontocetes	 4.162	 0.100		0.170	 16.000	 0.059	 0.011	12	 Seals	 4.265	 0.125		0.171	 15.100	 0.115	 0.011	13	 Sea	lions	 4.240	 0.125		0.171	 15.100	 0.082	 0.011	14	 Seabirds	 3.717	 0.007		0.100	 105.200	 0.033	 0.001	15	 Transient	salmon	 3.314	 0.208		2.480	 8.330	 0.817	 0.298	16	 Coho	salmon	 3.801	 0.024		2.760	 13.800	 0.607	 0.200	17	 Chinook	salmon	 3.748	 0.034		2.760	 13.800	 0.715	 0.200	18	 Small	squid	 2.939	 1.090		6.023	 34.675	 0.703	 0.174	19	 Large	squid	 3.028	 0.765		6.023	 34.675	 0.954	 0.174	20	 Octopus	 3.699	 0.190		0.860	 2.500	 0.900	 0.344	21	 Ratfish	 3.518	 0.517		0.099	 1.400	 0.792	 0.071	22	 Dogfish	 3.682	 0.909		0.099	 2.719	 0.739	 0.036	23	 Pollock	 3.443	 0.491		0.478	 2.280	 0.959	 0.209	24	 Forage	fish	 3.065	 8.478		1.600	 8.395	 0.996	 0.191	25	 Hake	 3.657	 0.820		0.550	 2.750	 0.957	 0.200	26	 Saury	 3.309	 1.303		1.600	 7.900	 0.950	 0.203	27	 Eulachon	 3.107	 1.660		1.432	 8.395	 0.882	 0.171		HG	herring		     		 		28	 HG	herring	age	0-1	yrs	 3.173	 0.048	 1.230		21.430	 0.948	 0.057	29	 HG	herring	age	1-3	yrs	 3.368	 0.475	 0.995		9.196	 0.970	 0.108	30	 HG	herring	age	3+	yrs	 3.514	 0.312	 1.057		5.840	 0.969	 0.181		HG_minor	herring		     		 		31	 HG_minor	age	0-1yrs	 3.173	 0.000	 1.230		23.665	 0.971	 0.052	32	 HG_minor	age	1-3	yrs	 3.368	 0.003	 0.452		9.557	 0.931	 0.047	33	 HG_minor	age	3+	yrs	 3.514	 0.009	 0.452		5.840	 0.898	 0.077		PRD	herring		     		 		34	 PRD	herring	age	0-1yrs	 3.173	 0.022	 1.230		22.306	 0.857	 0.055	35	 PRD	herring	age	1-3	yrs	 3.368	 0.320	 0.630		9.183	 0.839	 0.069	36	 PRD	herringage	3+	yrs	 3.514	 0.469	 0.740		5.840	 0.936	 0.127		CC	herring		     		 		37	 CC	herring	age	0-1	yrs	 3.173	 0.028	 1.230		22.870	 0.955	 0.054	38	 CC	herring	age	1-3	yrs	 3.368	 0.460	 0.520		9.303	 0.895	 0.056	39	 CC	herring	age	3+	yrs	 3.514	 0.975	 0.600		5.840	 0.911	 0.103	40	 POP	 3.299	 0.623		0.197	 2.246	 0.917	 0.087	41	 Inshore	rockfish	 3.611	 0.100		0.190	 5.688	 0.825	 0.033	42	 Piscivorous	rockfish	 3.385	 0.660		0.060	 1.267	 0.881	 0.047	43	 Planktivorous	rockfish	 3.458	 1.343		0.100	 2.248	 0.971	 0.044	A Revised EwE Model of Haida Gwaii 		 67	SN	 Group	name	 TL	 B	(t/km²)	 Z	(/year)	 P/B	(/year)	 Q/B	(/year)	 EE	 P/Q	(/year)	44	 Arrowtooth	flounder	 3.810	 1.748		0.234	 2.007	 0.996	 0.116	45	 Flatfish	 3.215	 0.495		1.465	 5.187	 0.704	 0.282		halibut		     		 		46	 Juvenile	halibut	 3.928	 0.356	 0.500		2.434	 0.984	 0.205	47	 Adult	halibut	 3.961	 0.900	 0.400		1.095	 0.546	 0.365	48	 Pacific	cod	 3.434	 0.252		1.553	 5.236	 0.976	 0.297	49	 Sablefish	 3.561	 0.388		0.375	 4.733	 0.864	 0.079	50	 Lingcod	 4.273	 0.070		0.977	 3.300	 0.871	 0.296	51	 Shallowwater	benthic	fish	 3.501	 0.509		1.500	 5.256	 0.994	 0.285	52	 Small	demersal	elasmobranchs	 3.641	 0.300		0.320	 1.240	 0.959	 0.258	53	 Large	demersal	sharks	 3.911	 0.025		0.130	 1.240	 0.048	 0.105	54	 Salmon	sharks	 4.385	 0.020		0.200	 1.200	 0.000	 0.167	55	 Blue	sharks	 4.008	 0.020		0.170	 0.800	 0.000	 0.213	56	 Large	crabs	 2.981	 0.456		1.500	 5.000	 0.956	 0.300	57	 Small	crabs	 3.154	 0.650		3.500	 14.000	 0.780	 0.250	58	 Commercial	shrimp	 2.719	 0.200		11.475	 45.900	 0.397	 0.250	59	 Sea	urchins	 2.000	 0.219		0.500	 10.880	 0.900	 0.046	60	 Other	grazers	 2.000	 11.469		0.753	 8.859	 0.900	 0.085	61	 Epifaunal	filter-feeders	 2.211	 9.738		1.000	 4.500	 0.800	 0.222	62	 Epifaunal	carnivores	 3.081	 1.664		0.850	 7.500	 0.900	 0.113	63	 Infaunal	carnivorous	invertebrates	 2.060	 13.245		2.000	 22.222	 0.235	 0.090	64	 Infaunal	detritivorous	invertebrates	 2.000	 34.305		1.349	 14.989	 0.537	 0.090	65	 Carnivorous	jellyfish	 2.173	 3.000		18.000	 60.000	 0.703	 0.300	66	 Macrozooplankton	 2.640	 1.654		7.000	 35.000	 0.800	 0.200	67	 Amphipods	 2.269	 1.047		7.000	 35.000	 0.800	 0.200	68	 Euphausiids	 2.326	 10.000		6.600	 24.820	 0.895	 0.266	69	 Copepods	and	cladocerans	 2.105	 5.250		27.000	 90.000	 0.986	 0.300	70	 Small	gelatinous	zooplankton	 2.466	 1.472		9.000	 30.000	 0.800	 0.300	71	 Microzooplankton	 2.053	 1.503		100.000	 285.714	 0.800	 0.350	72	 Corals	and	sponges	 2.000	 1.929		0.010	 2.000	 0.104	 0.005	73	 Eelgrass	 1.000	 1.242		24.542	 0.000	 0.400	 		74	 Kelps	 1.000	 0.215		15.000	 0.000	 0.400	 		75	 Benthic	macroalgae	 1.000	 3.977		15.000	 0.000	 0.400	 		76	 Benthic	microalgae	 1.000	 0.924		100.000	 0.000	 0.500	 		77	 Phytoplankton	 1.000	 15.406		178.502	 0.000	 0.380	 		78	 Detritus	 1.000	 10.000		   0.441	 				 	2016  Fisheries Centre Research Reports 24(2) 		68	Appendix B. Revised HG model: diet matrix   Prey \ predator 1 2 3 4 5 6 7 8 9 10 11 12 1 Sea Otters            2 Gray whales         0.012   3 Humpback whales            4 Minke whales         0.012   5 Blue whales            6 Fin whales            7 Sei whales            8 Sperm whales            9 Resident orcas            10 Transient orcas            11 Small odontocetes         0.075   12 Seals          0.100   13 Sea lions          0.051   14 Seabirds             15 Transient salmon        0.010  0.010 0.010 16 Coho salmon        0.010  0.003 0.001 17 Chinook salmon        0.380  0.005 0.004 18 Small squid 0.001       0.055   0.180 0.100 19 Large squid 0.001       0.375   0.112 0.030 20 Octopus 0.010           0.005 21 Ratfish 0.001       0.001     22 Dogfish        0.005   0.008 0.001 23 Pollock 0.001      0.013    0.025 0.002 24 Forage fish  0.100 0.090 0.005 0.040 0.025    0.364 0.353 25 Hake       0.008 0.005  0.020 0.010 26 Saury   0.015 0.001  0.035 0.030     0.030 27 Eulachon   0.015 0.009       0.042 0.030 28 HG herring age 0-1 yrs 0.001 0.001  0.000     0.007 0.006 29 HG herring age 1-3 yrs 0.015 0.012  0.005     0.070 0.064 30 HG herring age 3+ yrs 0.011 0.009  0.004     0.030 0.020 31 HG_minor age 0-1yrs 0.000 0.000  0.000     0.000 0.000 32 HG_minor age 1-3 yrs 0.000 0.000  0.000     0.001 0.000 33 HG_minor age 3+ yrs 0.000 0.000  0.000     0.001 0.000 34 PRD herring age 0-1yrs 0.000 0.000  0.000     0.001 0.004 35 PRD herring age 1-3 yrs 0.009 0.008  0.003     0.010 0.020 36 PRD herringage 3+ yrs 0.030 0.025  0.009     0.010 0.010 37 CC herring age 0-1 yrs 0.000 0.000  0.000     0.001 0.001 38 CC herring age 1-3 yrs 0.011 0.009  0.003     0.028 0.019 39 CC herring age 3+ yrs 0.040 0.034  0.013     0.010 0.001 40 POP           0.002 0.001 41 Inshore rockfish           0.001 42 Piscivorous rockfish       0.020   0.001 0.001 43 Planktivorous rockfish      0.020   0.005 0.002 44 Arrowtooth flounder      0.001   0.005 0.025 45 Flatfish        0.001   0.003 0.030 46 Juvenile halibut          0.003 0.010 47 Adult halibut       0.001   0.005 0.040 48 Pacific cod          0.025 0.061 49 Sablefish        0.005   0.010  50 Lingcod        0.002   0.001 0.00551 Shallowwater benthic fish 0.010          0.002 0.100 52 Small demersal elasmobranchs     0.008   0.001 0.001 53 Large demersal sharks      0.002     54 Salmon sharks            55 Blue sharks            56 Large crabs 0.075           0.002 57 Small crabs 0.030            58 Commercial shrimp            59 Sea urchins 0.751            60 Other grazers 0.030            61 Epifaunal filter-feeders 0.050            62 Epifaunal carnivores 0.030            63 Infaunal carnivorous invertebrates 0.075          64 Infaunal detritivorous invertebrates 0.010 0.400           65 Carnivorous jellyfish           66 Macrozooplankton            67 Amphipods            68 Euphausiids 0.025 0.652 0.501 0.475 0.525 0.200      69 Copepods and cladocerans   0.020 0.113 0.225      70 Small gelatinous zooplankton           71 Microzooplankton            72 Corals and sponges            73 Eelgrass             74 Kelps             75 Benthic macroalgae            76 Benthic microalgae            77 Phytoplankton            78 Detritus             A Revised EwE Model of Haida Gwaii 		 69	79 Import  0.500 0.100 0.300 0.500 0.250 0.500 0.500 0.600 0.750    Prey \ predator 13 14 15 16 17 18 19 20 21 22 23 24 1 Sea Otters            2 Gray whales            3 Humpback whales            4 Minke whales            5 Blue whales            6 Fin whales            7 Sei whales            8 Sperm whales            9 Resident orcas            10 Transient orcas            11 Small odontocetes            12 Seals             13 Sea lions             14 Seabirds             15 Transient salmon 0.100 0.020        0.010   16 Coho salmon 0.003         0.001   17 Chinook salmon 0.005         0.003   18 Small squid 0.090 0.050 0.200   0.100   0.066 0.006 19 Large squid 0.057 0.045  0.190   0.100   0.110 0.006  20 Octopus        0.050 0.030   21 Ratfish        0.030  0.002   22 Dogfish 0.001            23 Pollock 0.025          0.028  24 Forage fish 0.377 0.158 0.160 0.400 0.070 0.100  0.278 0.055 0.120  25 Hake 0.020         0.010 0.019  26 Saury 0.020 0.050 0.030 0.007 0.010 0.011 0.050   0.005 0.005  27 Eulachon 0.035 0.079  0.033 0.060 0.014 0.025  0.056 0.015 0.019  28 HG herring age 0-1 yrs 0.001 0.001  0.000 0.001     0.001   29 HG herring age 1-3 yrs 0.015 0.014  0.005 0.006     0.004   30 HG herring age 3+ yrs 0.012 0.010  0.004 0.005     0.006   31 HG_minor age 0-1yrs 0.000 0.000  0.000 0.000     0.000   32 HG_minor age 1-3 yrs 0.000 0.000  0.000 0.000     0.000   33 HG_minor age 3+ yrs 0.000 0.000  0.000 0.000     0.000   34 PRD herring age 0-1yrs 0.002 0.000  0.000 0.000     0.003   35 PRD herring age 1-3 yrs 0.010 0.009  0.003 0.004     0.004   36 PRD herringage 3+ yrs 0.010 0.028  0.010 0.013     0.008   37 CC herring age 0-1 yrs 0.001 0.000  0.000 0.001     0.000   38 CC herring age 1-3 yrs 0.011 0.010  0.004 0.005     0.003   39 CC herring age 3+ yrs 0.042 0.037  0.013 0.017     0.011   40 POP 0.005         0.004   41 Inshore rockfish 0.001            42 Piscivorous rockfish 0.001            43 Planktivorous rockfish 0.006         0.001   44 Arrowtooth flounder 0.030         0.015   45 Flatfish 0.030       0.010  0.040   46 Juvenile halibut 0.010            47 Adult halibut 0.030            48 Pacific cod 0.020   0.010      0.010   49 Sablefish 0.005         0.009   50 Lingcod 0.003            51 Shallowwater benthic fish 0.020       0.020  0.062   52 Small demersal elasmobranchs 0.002       0.010     53 Large demersal sharks           54 Salmon sharks            55 Blue sharks            56 Large crabs 0.002       0.140  0.037   57 Small crabs 0.041     0.140  0.026   58 Commercial shrimp       0.050     59 Sea urchins        0.001 0.001 0.002  60 Other grazers 0.010      0.150 0.049 0.020 0.020  61 Epifaunal filter-feeders 0.020      0.150 0.070  0.020  62 Epifaunal carnivores 0.010      0.130 0.060 0.020 0.011  63 Infaunal carnivorous invertebrates     0.050 0.200 0.019 0.013  64 Infaunal detritivorous invertebrates     0.070 0.080 0.009   65 Carnivorous jellyfish 0.036 0.100   0.060 0.060   0.029  0.270 66 Macrozooplankton 0.050 0.030 0.030 0.030    0.002  0.070 0.040 67 Amphipods 0.010 0.010 0.010 0.010      0.030 0.010 68 Euphausiids 0.112 0.150 0.300 0.419 0.380 0.107  0.204 0.135 0.310 0.200 69 Copepods and cladocerans 0.158 0.150   0.149 0.041   0.135 0.322 0.300 70 Small gelatinous zooplankton 0.030 0.020 0.020       0.050 71 Microzooplankton            72 Corals and sponges            73 Eelgrass             74 Kelps             75 Benthic macroalgae            76 Benthic microalgae            77 Phytoplankton            78 Detritus  0.041    0.316 0.417   0.080  0.130 79 Import   0.500         2016  Fisheries Centre Research Reports 24(2) 		70	 Prey \ predator 25 26 27 28 29 30 31 32 33 34 35 36 1 Sea Otters            2 Gray whales            3 Humpback whales            4 Minke whales            5 Blue whales            6 Fin whales            7 Sei whales            8 Sperm whales            9 Resident orcas            10 Transient orcas            11 Small odontocetes            12 Seals             13 Sea lions             14 Seabirds             15 Transient salmon            16 Coho salmon            17 Chinook salmon            18 Small squid 0.020            19 Large squid            20 Octopus             21 Ratfish             22 Dogfish             23 Pollock 0.005            24 Forage fish 0.100 0.025  0.050 0.220  0.050 0.220  0.050 0.220 25 Hake 0.015            26 Saury 0.005            27 Eulachon 0.001            28 HG herring age 0-1 yrs 0.005            29 HG herring age 1-3 yrs 0.040            30 HG herring age 3+ yrs 0.039            31 HG_minor age 0-1yrs 0.000            32 HG_minor age 1-3 yrs 0.000            33 HG_minor age 3+ yrs 0.000            34 PRD herring age 0-1yrs 0.001            35 PRD herring age 1-3 yrs 0.010            36 PRD herringage 3+ yrs 0.010            37 CC herring age 0-1 yrs 0.011            38 CC herring age 1-3 yrs 0.013            39 CC herring age 3+ yrs 0.015            40 POP             41 Inshore rockfish            42 Piscivorous rockfish 0.001            43 Planktivorous rockfish           44 Arrowtooth flounder           45 Flatfish             46 Juvenile halibut            47 Adult halibut            48 Pacific cod            49 Sablefish             50 Lingcod             51 Shallowwater benthic fish           52 Small demersal elasmobranchs          53 Large demersal sharks           54 Salmon sharks            55 Blue sharks            56 Large crabs            57 Small crabs            58 Commercial shrimp 0.124            59 Sea urchins            60 Other grazers            61 Epifaunal filter-feeders           62 Epifaunal carnivores           63 Infaunal carnivorous invertebrates 0.025          64 Infaunal detritivorous invertebrates 0.050          65 Carnivorous jellyfish 0.200          66 Macrozooplankton 0.099 0.080  0.100 0.190 0.230 0.100 0.190 0.230 0.100 0.190 0.230 67 Amphipods 0.040 0.020  0.020 0.180 0.250 0.020 0.180 0.250 0.020 0.180 0.250 68 Euphausiids 0.347 0.050 0.100 0.050 0.380 0.150 0.050 0.380 0.150 0.050 0.380 0.150 69 Copepods and cladocerans 0.099 0.300 0.600 0.830 0.200 0.150 0.830 0.200 0.150 0.830 0.200 0.150 70 Small gelatinous zooplankton 0.100           71 Microzooplankton            72 Corals and sponges            73 Eelgrass             74 Kelps             75 Benthic macroalgae            76 Benthic microalgae            77 Phytoplankton            78 Detritus   0.025          79 Import  0.425            Prey \ predator 37 38 39 40 41 42 43 44 45 46 47 48 A Revised EwE Model of Haida Gwaii 		 71	1 Sea Otters            2 Gray whales            3 Humpback whales            4 Minke whales            5 Blue whales            6 Fin whales            7 Sei whales            8 Sperm whales            9 Resident orcas            10 Transient orcas            11 Small odontocetes            12 Seals             13 Sea lions             14 Seabirds             15 Transient salmon            16 Coho salmon            17 Chinook salmon            18 Small squid   0.010  0.069 0.086 0.151  0.032 0.064  19 Large squid   0.010  0.079 0.086 0.152  0.032 0.064  20 Octopus           0.010  21 Ratfish             22 Dogfish             23 Pollock        0.020     24 Forage fish 0.050 0.220  0.060 0.020 0.024 0.121 0.091 0.055 0.014 0.172 25 Hake      0.006       26 Saury             27 Eulachon     0.012 0.005 0.005 0.047 0.018 0.011 0.003 0.039 28 HG herring age 0-1 yrs   0.001  0.000 0.004  0.001 0.000 0.000 29 HG herring age 1-3 yrs   0.015  0.005 0.007  0.006 0.002 0.003 30 HG herring age 3+ yrs   0.011  0.004 0.006  0.005 0.002 0.002 31 HG_minor age 0-1yrs   0.000  0.000 0.000  0.000 0.000 0.000 32 HG_minor age 1-3 yrs   0.000  0.000 0.000  0.000 0.000 0.000 33 HG_minor age 3+ yrs   0.000  0.000 0.000  0.000 0.000 0.000 34 PRD herring age 0-1yrs   0.000  0.000 0.000  0.001 0.000 0.000 35 PRD herring age 1-3 yrs   0.009  0.003 0.005  0.004 0.002 0.002 36 PRD herringage 3+ yrs   0.030  0.010 0.004  0.013 0.005 0.007 37 CC herring age 0-1 yrs   0.000  0.000 0.000  0.000 0.000 0.000 38 CC herring age 1-3 yrs   0.011  0.004 0.006  0.005 0.002 0.002 39 CC herring age 3+ yrs   0.040  0.013 0.020  0.017 0.007 0.009 40 POP      0.010 0.002  0.003 0.005 0.000 41 Inshore rockfish            42 Piscivorous rockfish            43 Planktivorous rockfish      0.002   0.006  44 Arrowtooth flounder      0.001  0.010 0.090 0.03045 Flatfish        0.013  0.020 0.103  46 Juvenile halibut          0.060 0.01047 Adult halibut            48 Pacific cod       0.003  0.008 0.050 0.010 49 Sablefish            0.002 50 Lingcod        0.002 0.002  0.005  51 Shallowwater benthic fish   0.045   0.050  0.055 0.06052 Small demersal elasmobranchs   0.020    0.020 0.005 53 Large demersal sharks           54 Salmon sharks            55 Blue sharks            56 Large crabs    0.005   0.011  0.200 0.153  57 Small crabs    0.107 0.130 0.011 0.056 0.348 0.154 0.01458 Commercial shrimp    0.095 0.012  0.057  0.060   59 Sea urchins     0.001   0.001 0.001 0.001 0.001 60 Other grazers    0.015 0.050  0.010 0.015 0.010 0.020 0.060 61 Epifaunal filter-feeders   0.015 0.050  0.010 0.020 0.010 0.020 0.060 62 Epifaunal carnivores   0.020 0.050  0.009 0.001 0.010 0.020 0.030 63 Infaunal carnivorous invertebrates  0.355 0.059   0.274 0.010  0.075 64 Infaunal detritivorous invertebrates   0.187   0.400   0.173 65 Carnivorous jellyfish    0.037 0.005     66 Macrozooplankton 0.100 0.190 0.230     0.100   0.020 67 Amphipods 0.020 0.180 0.250     0.020    0.005 68 Euphausiids 0.050 0.380 0.150 0.795 0.152 0.101 0.544 0.100 0.002  0.058 69 Copepods and cladocerans 0.830 0.200 0.150 0.185  0.005 0.190     0.029 70 Small gelatinous zooplankton           71 Microzooplankton      0.020  0.120    72 Corals and sponges            73 Eelgrass             74 Kelps             75 Benthic macroalgae            76 Benthic microalgae            77 Phytoplankton            78 Detritus      0.110  0.057  0.075 0.118 0.121 79 Import              Prey \ predator 49 50 51 52 53 54 55 56 57 58 59 60 1 Sea Otters            2016  Fisheries Centre Research Reports 24(2) 		72	2 Gray whales            3 Humpback whales            4 Minke whales            5 Blue whales            6 Fin whales            7 Sei whales            8 Sperm whales            9 Resident orcas            10 Transient orcas            11 Small odontocetes            12 Seals     0.034        13 Sea lions     0.034        14 Seabirds      0.001      15 Transient salmon     0.125 0.013     16 Coho salmon     0.015 0.003      17 Chinook salmon     0.015 0.003      18 Small squid 0.018  0.001  0.050 0.003 0.121      19 Large squid 0.018  0.001  0.200 0.004 0.049      20 Octopus  0.040 0.050 0.060        21 Ratfish    0.020 0.070 0.010      22 Dogfish     0.030 0.002       23 Pollock 0.005 0.011   0.010 0.001       24 Forage fish 0.180 0.140 0.198 0.040  0.025 0.013     25 Hake  0.026    0.002 0.013      26 Saury      0.010 0.008      27 Eulachon 0.039 0.063 0.040 0.010  0.001       28 HG herring age 0-1 yrs 0.000 0.002     0.000     29 HG herring age 1-3 yrs 0.001 0.018     0.001      30 HG herring age 3+ yrs 0.001 0.013     0.001      31 HG_minor age 0-1yrs 0.000 0.000     0.000      32 HG_minor age 1-3 yrs 0.000 0.000     0.000      33 HG_minor age 3+ yrs 0.000 0.000     0.000      34 PRD herring age 0-1yrs 0.000 0.001     0.000      35 PRD herring age 1-3 yrs 0.001 0.011     0.000      36 PRD herringage 3+ yrs 0.003 0.036    0.001 0.002      37 CC herring age 0-1 yrs 0.000 0.001     0.000      38 CC herring age 1-3 yrs 0.001 0.013     0.001      39 CC herring age 3+ yrs 0.003 0.048    0.001 0.002      40 POP  0.012           41 Inshore rockfish 0.006   0.001        42 Piscivorous rockfish 0.001   0.001 0.001      43 Planktivorous rockfish 0.003 0.011   0.010 0.001       44 Arrowtooth flounder 0.010 0.011   0.050 0.002       45 Flatfish 0.005 0.093  0.028 0.020 0.001  0.010     46 Juvenile halibut 0.003 0.099   0.010        47 Adult halibut 0.028   0.040        48 Pacific cod 0.003 0.089   0.010 0.001      49 Sablefish 0.016    0.050 0.028       50 Lingcod  0.010  0.010        51 Shallowwater benthic fish 0.010 0.028 0.010        52 Small demersal elasmobranchs  0.020        53 Large demersal sharks           54 Salmon sharks            55 Blue sharks            56 Large crabs   0.250 0.061        57 Small crabs 0.006 0.076 0.181 0.125    0.100     58 Commercial shrimp 0.079 0.081 0.004 0.125    0.015     59 Sea urchins   0.005    0.001     60 Other grazers 0.005 0.010 0.010 0.005    0.137 0.260   61 Epifaunal filter-feeders 0.005 0.010 0.020 0.015    0.137 0.260    62 Epifaunal carnivores 0.005 0.006 0.007 0.005    0.100 0.075    63 Infaunal carnivorous invertebrates 0.333 0.150    0.100 0.305    64 Infaunal detritivorous invertebrates 0.096 0.045    0.100 0.100 0.100  65 Carnivorous jellyfish 0.188            66 Macrozooplankton 0.015          67 Amphipods 0.005           68 Euphausiids 0.382  0.025   0.022   0.300   69 Copepods and cladocerans 0.048       0.200   70 Small gelatinous zooplankton           71 Microzooplankton            72 Corals and sponges            73 Eelgrass            0.120 74 Kelps           0.500 0.001 75 Benthic macroalgae          0.250 0.229 76 Benthic microalgae          0.200 0.450 77 Phytoplankton            78 Detritus 0.021 0.016 0.027 0.100 0.220   0.300  0.400 0.050 0.200 79 Import      0.750 0.750       Prey \ predator 61 62 63 64 65 66 67 68 69 70 71 72 1 Sea Otters            2 Gray whales            A Revised EwE Model of Haida Gwaii 		 73	3 Humpback whales            4 Minke whales            5 Blue whales            6 Fin whales            7 Sei whales            8 Sperm whales            9 Resident orcas            10 Transient orcas            11 Small odontocetes            12 Seals             13 Sea lions             14 Seabirds             15 Transient salmon            16 Coho salmon            17 Chinook salmon            18 Small squid            19 Large squid            20 Octopus             21 Ratfish             22 Dogfish             23 Pollock             24 Forage fish            25 Hake             26 Saury             27 Eulachon             28 HG herring age 0-1 yrs           29 HG herring age 1-3 yrs           30 HG herring age 3+ yrs           31 HG_minor age 0-1yrs           32 HG_minor age 1-3 yrs           33 HG_minor age 3+ yrs           34 PRD herring age 0-1yrs           35 PRD herring age 1-3 yrs           36 PRD herringage 3+ yrs           37 CC herring age 0-1 yrs           38 CC herring age 1-3 yrs           39 CC herring age 3+ yrs           40 POP             41 Inshore rockfish            42 Piscivorous rockfish            43 Planktivorous rockfish           44 Arrowtooth flounder           45 Flatfish             46 Juvenile halibut            47 Adult halibut            48 Pacific cod            49 Sablefish             50 Lingcod             51 Shallowwater benthic fish           52 Small demersal elasmobranchs          53 Large demersal sharks           54 Salmon sharks            55 Blue sharks            56 Large crabs            57 Small crabs            58 Commercial shrimp            59 Sea urchins 0.001           60 Other grazers 0.372           61 Epifaunal filter-feeders 0.372           62 Epifaunal carnivores           63 Infaunal carnivorous invertebrates 0.050           64 Infaunal detritivorous invertebrates 0.206 0.060         65 Carnivorous jellyfish   0.060 0.010    0.001   66 Macrozooplankton            67 Amphipods            68 Euphausiids    0.010 0.100    0.050   69 Copepods and cladocerans   0.081 0.300  0.200  0.200   70 Small gelatinous zooplankton    0.040 0.040  0.049   71 Microzooplankton 0.200     0.100 0.200 0.100 0.100 0.100 0.050 72 Corals and sponges            73 Eelgrass             74 Kelps             75 Benthic macroalgae            76 Benthic microalgae            77 Phytoplankton 0.400     0.300 0.260 0.700 0.900 0.350 0.900  78 Detritus 0.400  0.940 1.000 0.849 0.150 0.500   0.250 0.050 1.00079 Import             	2016  Fisheries Centre Research Reports 24(2) 		74	Appendix C: Revised HG model: fisheries (kg.km-2) Fleets/Groups 15 16 17 22 23 25 29 30 35 36 38 39 40 41 42 43 Groundfish trawl    0.7 6.7        59.9 0.3 22.6 76.5 Sable                 Herring gillnet         0.3 37.6 0.1 11.9     Ground H+L              3.0 2.0  Salmon gillnet 75.0 20.0 15.0              Crab trap                 Shrimp / prawn trap                 Other Inv.                 Halibut H+L              4.0   Salmon troll 2.5 0.7 0.3              Salmon seine 66.7 0.2               Salmon troll freezer 9.7 1.4 0.5              Herring seine       1.1 20.7 0.1 15.2 0.4 78.5     Shrimp trawl                 Longline    30.0             Recreational 0.7 2.3 9.3           3.0 2.0  Hake      252.0           HG salmon 0.1 0.1               HG_herring SOK        11.2  12.4  19.1     HG Clam                 HG Seaweed                 HG_FSC                 Fleets/Groups 44 45 46 47 48 49 50 52 56 58 59 60 61 62 64 75 Groundfish trawl 67.0 58.0 0.1 0.1 51.0 0.4 19.0 28.7   0.0 1.0 0.8 0.1   Sable      38.3           Herring gillnet                 Ground H+L                 Salmon gillnet                 Crab trap         25.6        Shrimp / prawn trap          3.8       Other Inv.           62.3    15.7  Halibut H+L   27.8 30.7 2.8 2.1           Salmon troll                 Salmon seine                 Salmon troll freezer                 Herring seine                 Shrimp trawl          32.7       Longline                 Recreational   1.4 15.5   2.8  1.0 0.2       Hake                 HG salmon                 HG_herring SOK                 HG Clam             1.3 1.3   HG Seaweed                0.0 HG_FSC                 		 	A Revised EwE Model of Haida Gwaii 		 75	Appendix D. R code for extracting relevant data from ArcGIS files for Ecospace maps #lower	left	corner	for	HG	hgx=479233.353991		hgy=714688.1922				#rows	and	columns	in	Ecospace	map	ewecol=60		ewerow=97		#lower	left	corner	of	ArcGIS	file	asciix=459060	#easting	corresponds	to	lower	left	longitude	#Change	here	asciiy=539650	#northing	corresponds	to	lower	left	latitude	#	Change	here	#my_ascii=read.table("C:/Dropbox/Spatial	layers/marxan2_mpa.txt",skip=6)		my_ascii=read.table("marxan2_mpa.txt",skip=6)#change	here	(external	file	name)	out_name="mpa_2.csv"																#output	file	name		#to	check	the	top	left	boundary	indx=(asciix-hgx)/4000	#if	indx<1,	then	left	boundary	of	shape	file	is		#to	the	left	of	the	modelled	map	area	as	in	panel	a	or	c	in	Figure	12	#indx>1	means	that	some	parts	of	modelled	area	are	missing	as	#	in	panel	b	or	d	of	Figure	12	indx=round(indx,0)		indy=(asciiy-hgy)/4000	#if	indy<1,	then	top	boundary	of	shape	file	is		#below	the	modelled	map	area	as	in	panel	c	or	d	in	Figure	12	#indy>1	means	that	some	parts	of	modelled	area	are	missing	as	#in	panel	a	of	Figure	12,	indy	=	0	as	in	panel	b	of	Figure	12.	indy=round(indy,0)		dim_as=dim(my_ascii)	#dim	function:	dimension	of	a	matrix:	row	and	col		if	(	indx>=1	)		{			new_col=matrix(data=-99,nrow=dim_as[1],ncol=indx)			new_ascii=cbind(new_col,my_ascii)	#extra	blank	columns	are	added	to	the	left	}	else	if	(	indx<=-1	)		{			indxx=seq(1,abs(indx))					new_ascii=my_ascii[,-indxx]	#extra	columns	outside	the	map	area	are	removed	from	the	left	}	else	new_ascii=my_ascii		dim_nas=dim(new_ascii)	if	(	indy>=1	)		{	2016  Fisheries Centre Research Reports 24(2) 		76			new_row=matrix(data=-99,nrow=abs(indy),ncol=dim_nas[2])			colnames(new_row)<-colnames(new_ascii)			new_ascii2=rbind(new_ascii,new_row)	#extra	blank	columns	are	added	on	the	lower	edge	}	else	if	(	indy<=-1	)		{			indyy=seq(dim_nas[1],dim_nas[1]-(abs(indy)-1))			new_ascii2=new_ascii[-indyy,]	#extra	columns	are	removed	from	lower	edge	}	else	new_ascii2=new_ascii		#########################################################	#THIS	SECTION	NOW	CUTS	THE	MAP	TO	HAIDA	GWAII	MODELLED	AREA	#########################################################	dim_nas2=dim(new_ascii2)	inde=dim_nas2[1]-ewerow	#calculates	extra	columns	east	of	HG	area	if	(	inde>=1	)		{			xx=seq(dim_nas2[1]-(ewerow-1),dim_nas2[1])			new_ascii3=new_ascii2[xx,]	}	else	if	(	inde<=-1	)	#adds	extra	columns	to	complete	the	east	boundary	{			new_row2=matrix(data=-99,nrow=abs(inde),ncol=dim_nas2[2])			colnames(new_row2)<-colnames(new_ascii2)			new_ascii3=rbind(new_row2,new_ascii2)	}	else	new_ascii3=new_ascii2			dim_nas3=dim(new_ascii3)		indf=dim_nas3[2]-ewecol	#calculates	extra	rows	north	of	HG	area		if	(	indf>=1	)	#trims	to	north	HG	boundary	{			yy=seq(1:ewecol)			new_ascii4=new_ascii3[,yy]		}	else	if	(	indf<=-1	)		#adds	extra	rows	to	complete	the	north	bundary	{			new_col2=matrix(data=-99,nrow=dim_nas3[1],ncol=abs(indf))			new_ascii4=cbind(new_ascii3,new_col2)	}	else	new_ascii4=new_ascii3		print(dim(new_ascii4))	#map	for	Ecospace	write.table(new_ascii4,out_name,	row.names=F,	col.names=F)		

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.52383.1-0354480/manifest

Comment

Related Items