CERF Mar<strong>in</strong>e Biodiversity Hub <strong>Jervis</strong> <strong>Bay</strong> Survey Report, February 2009map patterns <strong>of</strong> biodiversity through <strong>the</strong> use <strong>of</strong> physical surrogates, wherebyenvironmental variables are used to predict patterns <strong>of</strong> mar<strong>in</strong>e biodiversity.Field surveys are currently be<strong>in</strong>g undertaken <strong>in</strong> <strong>the</strong> Surrogates Program <strong>of</strong> <strong>the</strong> CERFMar<strong>in</strong>e Biodiversity Hub to provide data that will enable a range <strong>of</strong> physical factors to betested as useful proxies for mar<strong>in</strong>e benthic biodiversity. In <strong>the</strong> context <strong>of</strong> this program,<strong>Jervis</strong> <strong>Bay</strong> has been selected to test surrogates for <strong>the</strong> distribution <strong>of</strong> s<strong>of</strong>t-sediment benthicspecies <strong>and</strong> communities. The ma<strong>in</strong> datasets be<strong>in</strong>g collected are multibeam sonarbathymetry, sediment samples, oceanographic measurements, <strong>in</strong>fauna samples <strong>and</strong> towedunderwater video (Anderson et al. 2009).Environmental variables can affect <strong>in</strong>vertebrate abundance <strong>and</strong> species richness (Radke etal. 2003), although <strong>the</strong> strength <strong>of</strong> such effects will be species-specific <strong>and</strong> vary acrossfeed<strong>in</strong>g modes <strong>and</strong> habitat requirements. Compared to epibenthic <strong>in</strong>vertebrates <strong>and</strong>demersal fish, <strong>in</strong>faunal <strong>in</strong>vertebrates are most likely to be strongly related to benthicfactors such as sediment gra<strong>in</strong> size <strong>and</strong> organic content, because <strong>the</strong>y are directly exposedto <strong>the</strong>se factors via burrow<strong>in</strong>g <strong>and</strong> feed<strong>in</strong>g.Surveys designed to <strong>in</strong>vestigate appropriate surrogates normally make a trade-<strong>of</strong>f <strong>in</strong> <strong>the</strong>irsampl<strong>in</strong>g plan between overall spatial coverage <strong>and</strong> replication. Larger spatial coverage<strong>of</strong>ten means less replication or <strong>f<strong>in</strong>e</strong>-<strong>scale</strong> sampl<strong>in</strong>g. F<strong>in</strong>e-<strong>scale</strong> sampl<strong>in</strong>g may beparticularly important <strong>in</strong> heterogeneous habitats <strong>and</strong> communities, where a s<strong>in</strong>gle sampleis unable to encompass <strong>the</strong> <strong>variation</strong> at that station <strong>and</strong> is <strong>the</strong>refore not representative(Archambault & Bourget 1996, Wahl 2001). S<strong>of</strong>t sediment habitats are <strong>of</strong>ten consideredhomogenous habitats because <strong>the</strong>y appear this way us<strong>in</strong>g common sampl<strong>in</strong>g methods thatonly focus on <strong>the</strong> seabed surface (e.g. video, multibeam, epibenthic sleds). However, s<strong>of</strong>tsedimenthabitats can also be heterogeneous, with organic enrichment, oxygenation, <strong>and</strong>gra<strong>in</strong> size sometimes vary<strong>in</strong>g across meters or less (Kelaher & Lev<strong>in</strong>ton 2003, Wodarska-Kowaczuk & Weslawski 2008). Indeed, a study <strong>of</strong> <strong>in</strong>faunal assemblages <strong>in</strong> <strong>Jervis</strong> <strong>Bay</strong>showed high levels <strong>of</strong> heterogeneity at <strong>f<strong>in</strong>e</strong> <strong>scale</strong>s, with assemblages vary<strong>in</strong>g greatlyamong replicates (Hutch<strong>in</strong>gs & Jacoby 1994). It is unknown if all s<strong>of</strong>t sediment habitats<strong>in</strong> <strong>Jervis</strong> <strong>Bay</strong> show this level <strong>of</strong> heterogeneity or if abiotic factors similarly vary across<strong>f<strong>in</strong>e</strong> spatial <strong>scale</strong>s. None<strong>the</strong>less, it is generally accepted that <strong>variation</strong> is greater acrosslarger spatial <strong>scale</strong>s (hundreds <strong>of</strong> meters) than <strong>f<strong>in</strong>e</strong>r spatial <strong>scale</strong>s (tens <strong>of</strong> meters), likelydue to differences <strong>in</strong> habitat types.The physical data upon which surrogacy research relies is <strong>of</strong>ten pooled from multipletimes, sometimes spann<strong>in</strong>g several seasons. Abiotic factors such as temperature, organicloads, <strong>and</strong> light attenuation vary across seasons (Holloway 1995, Przeslawski & Davis2007); <strong>and</strong> identification <strong>of</strong> <strong>the</strong>se factors as potential surrogates may <strong>the</strong>refore beh<strong>in</strong>dered if sampl<strong>in</strong>g times are pooled across weeks, months, or seasons.Although spatial <strong>and</strong> temporal <strong>variation</strong> have been well-studied <strong>in</strong> seagrass <strong>and</strong> fishcommunities <strong>of</strong> <strong>Jervis</strong> <strong>Bay</strong> (Hutch<strong>in</strong>gs & Jacoby 1994, Jacoby et al. 1995a), <strong>the</strong>re havebeen comparatively few similar studies on <strong>in</strong>vertebrate assemblages <strong>in</strong> deeper waters(Hutch<strong>in</strong>gs & Jacoby 1994) <strong>and</strong> none on spatio-temporal <strong>variation</strong> <strong>of</strong> seafloorgeochemistry <strong>and</strong> <strong>biogeochemistry</strong>.2
CERF Mar<strong>in</strong>e Biodiversity Hub <strong>Jervis</strong> <strong>Bay</strong> Survey Report, February 20091.3 JERVIS BAY<strong>Jervis</strong> <strong>Bay</strong> is a temperate mar<strong>in</strong>e environment on <strong>the</strong> south coast <strong>of</strong> NSW (35°08’S,150°45’E), valued both economically <strong>and</strong> ecologically (Figure 1). The region is used for aunique comb<strong>in</strong>ation <strong>of</strong> recreational, commercial, <strong>and</strong> military purposes, <strong>and</strong> <strong>the</strong>ecosystem is <strong>the</strong>refore important to a broad range <strong>of</strong> people <strong>and</strong> groups. <strong>Jervis</strong> <strong>Bay</strong>conta<strong>in</strong>s several types <strong>of</strong> habitats, <strong>in</strong>clud<strong>in</strong>g seagrass, s<strong>and</strong>, mud, <strong>and</strong> rocky bottoms(Figure 1); <strong>and</strong> it receives little freshwater <strong>in</strong>put, <strong>in</strong> contrast to o<strong>the</strong>r embayments <strong>in</strong> <strong>the</strong>region (Ward 1995). The sou<strong>the</strong>rn waters <strong>of</strong> <strong>Jervis</strong> <strong>Bay</strong> are part <strong>of</strong> <strong>the</strong> National BoodereePark, <strong>and</strong> <strong>the</strong> rema<strong>in</strong><strong>in</strong>g waters encompass <strong>the</strong> <strong>Jervis</strong> <strong>Bay</strong> Mar<strong>in</strong>e Park, <strong>in</strong>clud<strong>in</strong>g HabitatProtection <strong>and</strong> Sanctuary Zones. In <strong>the</strong> rest <strong>of</strong> this section, we briefly review <strong>in</strong>formationrelated to <strong>variation</strong> <strong>of</strong> biogeochemical factors <strong>in</strong> <strong>Jervis</strong> <strong>Bay</strong>, specifically what is knownabout <strong>the</strong> biology <strong>and</strong> geochemistry <strong>of</strong> <strong>Jervis</strong> <strong>Bay</strong>, as well as spatial <strong>and</strong> temporal<strong>variation</strong> among biological <strong>and</strong> physical factors <strong>in</strong> this region.Figure 1: Map <strong>of</strong> <strong>Jervis</strong> <strong>Bay</strong> show<strong>in</strong>g Darl<strong>in</strong>g Road Grid outl<strong>in</strong>ed <strong>in</strong> black. Colours denote benthichabitats taken from <strong>Jervis</strong> <strong>Bay</strong> National Parks classifications based on historical data from mid-1980s.1.3.1. Benthic Biology <strong>of</strong> <strong>Jervis</strong> <strong>Bay</strong><strong>Jervis</strong> <strong>Bay</strong> is home to a mix <strong>of</strong> temperate <strong>and</strong> tropical species (Millar 1995), <strong>in</strong>clud<strong>in</strong>gnearly half <strong>of</strong> <strong>the</strong> total number <strong>of</strong> mar<strong>in</strong>e algae species <strong>in</strong> New South Wales, mak<strong>in</strong>g itone <strong>of</strong> <strong>the</strong> richest sites <strong>in</strong> eastern Australia for mar<strong>in</strong>e algae (Millar 1995). Four ma<strong>in</strong>biological assemblages are described <strong>in</strong> (Dames & Moore 1985) based on dom<strong>in</strong>antspecies: ‘heart urch<strong>in</strong> community’ with high densities <strong>of</strong> Ech<strong>in</strong>ocardium cordatum, <strong>the</strong>‘impoverished macrobenthic community’ with no obvious sessile organisms, <strong>the</strong>‘polychaete hummock community’ with chaetopterid worms <strong>and</strong> <strong>the</strong>ir mounds, <strong>and</strong> <strong>the</strong>‘mud oyster / mussel community’ with clusters <strong>of</strong> bivalves. This prelim<strong>in</strong>ary <strong>and</strong> coarse3