Aquatic Botany, 36 (1989) 69-77 69 Elsevier Science Publishers B.V., Amsterdam -- Printed in The Netherlands Effect of Boat Moorings on Seagrass Beds near Perth, Western Australia D.I. WALKER, R.J. LUKATELICH, G. BASTYAN and A.J. McCOMB Department of Botany and Centre for Water Research, University of Western Australia, Nedlands, W.A. 6009 (Australia) (Accepted for publication 3 July 1989 ) ABSTRACT Walker, D.I., Lukatelich, R.J., Bastyan, G. and McComb, A.J., 1989. Effect of boat moorings on seagrass beds near Perth, Western Australia. Aquat. Bot., 36: 69-77. Boat moorings have been found to produce circular scours in seagrass meadows, ranging from 3 to 300 m 2. "Cyclone" moorings (which have three anchors and a swivel ) are much less damaging to seagrass meadows than "swing" moorings (with a single anchor and chain). The total area of seagrass meadow lost due to moorings totals some 5.4 ha in the Rottnest Island, Warnbro Sound and Cockburn Sound regions of Western Australia, with most loss {3.14 ha) in the Rottnest region. While the relative area of seagrass meadow lost is small (< 2% ), there is considerable visual impact in some areas. The scours created by moorings in the seagrass canopy interfere with the physical integrity of the meadow. Though relatively small areas of seagrass are damaged by moorings, the effect is much greater than if an equivalent area was lost from the edge of a meadow. INTRODUCTION Loss of seagrass has been reported from all regions of the world, owing to apparently natural phenomena, e.g. "wasting disease" (den Hartog, 1987) or high energy storms (Patriquin, 1975), but more commonly as a consequence of light reduction, often as a result of eutrophication (Bulthuis, 1983; Orth and Moore, 1983; Cambridge and McComb, 1984; Neverauskas, 1987) or land reclamation and changes in land use (Kemp et al., 1983). However, in addition there are losses due to the construction of groynes and marinas, as well as smaller scale impacts, particularly from locally dense clusters of boat moorings. The Perth metropolitan region (Fig. 1) has extensive seagrass meadows, situated between the coast and a chain of offshore limestone reefs (Kirkman and Walker, 1989). These are mainly monospecific stands of Posidonia australis Hook. f., P. sinuosa Cambridge and Kuo, Amphibolis antarctica (Labill.) Sonder and Aschers., and smaller areas of A. griffithii (J.M. Black) den Hat- 0304-3770/89/$03.50 1989 Elsevier Science Publishers B.V.
.~ 70 D.I. WALKER ET AL. 0.0 4.0 Kilometres 8.0 * I Longreach Catherine Bay.,. ~ r d i e Bay Stark Bay "~" v {~t:,ql'hornson Bay Rocky Bay~ ~...,~,~'~ Marjorie Bay;m~" "~= ' '~2,/, c~f~ -" "'~Porpolse Bay INDIAN OCEAN q KEY Mooring Region Cockburn Sound WESTERN AUSTRALIA metropolitan ~, i' I Warnbro too,on.~ ~ Sou.. Fig. 1. The location of moorings within seagrass meadows along the Perth Metropolitan coast and Rottnest Island. tog. These meadows are interrupted by areas of sand ("blowouts") which are colonised by a variety of smaller species (Kirkman, 1985). Perth also has the highest per capita boat ownership in Australia (Western Australian Department of Marine and Harbours, personal communication, 1987) and boats are often left moored for long periods. Two types of moorings are commonly used in the region. A "cyclone" mooring is installed by blasting three radially arranged holes, into which anchor weights are inserted. These are connected by heavy chain to a central ring on a swivel, which is connected via another chain to a surface buoy. "Cyclone" moorings are generally installed by professional divers."swing" moorings (Fig.
EFFECT OF MOORINGS ON SEAGRASS BEDS IN AUSTRALIA 71 Fig. 2. Photograph of a swing mooring showing typical damage caused by the mooring chain to the seagrass meadow. The seagrass in the centre of the scoured area is drift material. {Photograph: G. Bastyan. ) 2) are more common and each consists of a single heavy anchor (often an engine block) connected to a surface buoy by a long length of heavy chain. This paper documents damage to seagrass meadows by boat moorings in relation to the total area of seagrass meadow in the region concerned, the potential recovery of the meadows after this disturbance, and effects of the damage on the physical and biological roles of seagrass meadows. METHODS Seagrass meadows were surveyed along the coast from Ocean Reef to Becher Point and offshore to Rottnest Island (Fig. 1). Locations of moorings were provided by the Department of Marine and Harbours. For convenience, the study area was divided into three main regions, Rottnest Island, Cockburn Sound and Warnbro Sound. For Rottnest Island, where all moorings are registered, the number and location of moorings was supplied by the Rottnest Island Board. The number of moorings in Cockburn and Warnbro Sounds was unknown and numbers were estimated from field surveys. The mean scour area per mooring in each region was calculated from measurements of scour diameter, made using a tape measure or passing a boat of known length over the scour. Measurements were made of 153 scours at Rottnest Island, 120 in Cockburn Sound and 66 in Warnbro Sound. In each of the
72 D.I. WALKER ET AL. mooring regions, the scoured area due to boat moorings was estimated by multiplying the mean scour area for that region by the number of moorings. The listed aerial photographs were also used to estimate areas of seagrass and seagrass loss. Western Australian Lands and Surveys Department: aerial photographs Photograph Date Area Run no. Scale series no. W.A. Dec. 1976 Cockburn and Warnbro Runs 1-11 1 : 10 000 W.A. 2235 June 1984 Cockburn Sound Runs 1-7 1 : 10 000 W.A. 2346 Sep. 1985 Mullaloo (5123-5130) 1:8 000 W.A. 1859 Feb. 1986 Rottnest Run 2 (5238) 1:10 000 W.A. 2490 Feb. 1987 Mullaloo-Mandurah (5239) 1 : 15 000 W.A. 2487 Feb. 1987 Mullaloo (5171-5179) 1 : 8 000 RESULTS A total of 1291 moorings was found in the study. These were all situated in dense monospecific seagrass meadows of Posidonia or Amphibolis spp. The area of seagrass meadow lost because of a single mooring ranged from 3 to 300 m 2, depending on the type of mooring and length of chain used. "Cyclone" moorings were found to cause a permanent scour of 3 m 2. "Swing" moorings scoured an area which depended on the length of chain used, usually a function of boat size. The mean size of the scoured area of swing moorings in Cockburn and Warnbro Sounds was found to be 39 m 2, but swing moorings for larger boats ( > 8 m) resulted in a much greater scoured area, 176-314 m 2. The floor of a scour was depressed below the sediment surface level of the surrounding meadow by ~ 0.5 m (Fig. 3), but scours as deep as 1 m below the surrounding sediment surface were observed at Rottnest Island. Areas of seagrass meadow lost due to boat moorings in different sections of the metropolitan region are shown in Table 1. Most damage was found around Rottnest Island (Fig. 1 ), which has large numbers of moorings (933). The area of seagrass lost due to boat moorings represents 1.4% of the total area of seagrass around Rottnest Island. In Thomson Bay, which has the largest number of moorings (344) of any bay on Rottnest Island, ~ 2.45 ha of seagrass meadow have been lost as a direct result of boat moorings (Fig. 4). Most moorings in Cockburn Sound are situated in Mangles Bay (Fig. 1 ). Of 253 boat moorings in this area, 102 are located outside the seagrass meadow and cause no damage to seagrass. The remaining moorings (151) have resulted in the loss of 1.8 ha of seagrass, which represents ~ 1.9% of meadow in this area. Most of the seagrass loss in Mangles Bay, ~ 1.4 ha, has been caused by two large barges (Fig. 5) which are moored in shallow water and probably sit on the bottom at low tide. There were large numbers of sand patches within the mooring region in Cockburn Sound that may have resulted from previous
EFFECT OF MOORINGS ON SEAGRASS BEDS IN AUSTRALIA 73 Fig. 3. Photograph showing a depression in the seagrass meadow caused by a swing mooring. Note intact seagrass meadow at the edge of the scoured area. (Photograph: G. Bastyan.) TABLE1 Area of seagrass meadows lost in the metropolitan region due to boat moorings Location Number of Scoured area Area of seagrass within Percentage of moorings (m 2 ) mooring region (m 2 ) seagrass meadow lost due to moorings Rottnest Island Catherine Bay 31 250 81 000 0.3 Stark Bay 105 1 830 160 000 1.1 Rocky Bay 191 2 390 319 000 0.8 Geordie Bay 92 720 105 000 0.7 Longreach Bay 73 590 102 000 0.6 Porpoise Bay 33 130 286 000 0.1 Marjorie Bay 64 1 060 245 000 0.4 Thomson Bay 344 24 380 934 400 2.6 Total (Rottnest) 933 31 350 2 232 400 1.4 Warnbro Sound 105 4 460 208 000 2.1 Cockburn Sound 253 18 360 955 000 1.9 Total 1 291 54 170 3 395 400 1.6
74 D.I. WALKER ET AL. Fig. 4. Aerial photograph showing damage to the seagrass meadow in Thomson Bay, Rottnest Island. (Photograph: Lands and Surveys Department.) Fig. 5. Aerial photograph showing damage to the seagrass meadow in Mangles Bay. Note the large scoured areas caused by the barges. (Photograph: S. Chape, State Planning Commission.)
EFFECT OF MOORINGS ON SEAGRASS BEDS IN AUSTRALIA 75 moorings, but these areas ( ~ 1.5 ha) were not included because of the uncertainty of their origin. In Warnbro Sound (Fig. 1 ), the moorings are generally placed between the shore and the 2.0-m depth contour, an area of extensive P. australis meadow. The area of seagrass within which the moorings are situated is ~ 21 ha, of which some 0.45 ha (2.1%) has been lost as a result of boat moorings currently in place. There were also numerous sand patches in this region which may have been scoured by moorings which have since been removed. This would increase the estimate of loss to 2.7 ha. The proportion of the total area of seagrass in Warnbro Sound (940 ha) is small (0.05%), but the mooring damage has a high visual impact. DISCUSSION The proportion of total seagrass meadow lost in the Perth Metropolitan region through boat mooring damage is quite small, 1.6% of 3.4 106 m 2, but reaches higher proportions in localised areas and may be visually striking. The estimate for Cockburn and Warnbro Sounds is most likely underestimated; as noted above, some scoured areas which may have been caused by moorings which had been temporarily or permanently removed were not included in the estimates. Although the loss is generally small, the scoured area results in an increase in the "edge" length which is vulnerable to erosion, e.g. mooring damage has resulted in an increase in the length of the eroding edge of seagrass of ~ 8.5 km in Cockburn and Warnbro Sounds, and 13.5 km around Rottnest Island. With up to 15 m 2 of leaf surface (McComb et al., 1981) lost for each m 2 of seafloor scoured of seagrass, the scours are differentially eroded and this can lead to undercutting of the surrounding area, with greater potential for erosion during subsequent storms. This undercutting may already have caused the loss of significant areas of seagrass; there were numerous 'blowouts' and thinning of the seagrass meadow within all the mooring regions. Seagrass detritus collects in these depressions and may restrict the chances of successful recolonisation. Loss of detritus and nutrients from the area damaged may extent further into the meadow as the edges are less effective at baffling the water movement. Seagrass meadows are important to near-shore coastal ecosystems, where they play significant roles in important geomo~phological and ecological processes, as they have both physical and biological effects on the environment. All these functions rely on the existence of an intact, well developed leaf canopy. Where the meadow becomes sparse or, more particularly in this case, if scours are created in the canopy by moorings (e.g. Figs. 2 and 3), the meadow ceases to have a physical integrity. Though only relatively small areas of seagrass are involved in mooring damage, the effect is much greater than if the equivalent area was lost from the edge of a meadow.
76 D.I. WALKER ET AL. Recolonisation by seagrass of previously damaged areas was observed in Porpoise Bay at Rottnest Island. About 10 moorings had been removed from Porpoise Bay over winter and after 6-8 months the scoured areas were found to be recolonised by pioneer seagrasses such as Halophila ovalis (R.Br.) Hook. f. and Heterozostera tasmanica (Martens ex Aschers.) den Hartog; however, these species are very susceptible to storm erosion (Kirkman, 1985). The moorings were located in a P. australis meadow and it is not known ifposidonia would re-establish in these scoured areas if the moorings were not replaced. Mining for P. australis fibre in Spencer Gulf, South Australia, in 1917 left scars which are still visible (Kirkman and Parker, 1978). In Botany Bay, New South Wales, experimental clearings of P. australis remained bare over a 35-month study period (Larkum, 1976). It seems unlikely that areas scoured of seagrass by moorings will be recolonised by meadow-forming species in the short term. The tidal range in the Perth Metropolitan region is < 1 m and so the length of swing moorings does not need to allow for large tidal fluctuations. Obviously, in areas with large tidal ranges, for example 8-10 m in the northwest of Western Australia, the length of mooring chains will be substantially greater and so cause greater erosion problems as a consequence of both chain length and tidal movements, leading to greater scour. Swing moorings are clearly more damaging than cyclone moorings, causing scours more than 10 times the area of those created by cyclone moorings. Damage resulting from moorings could be minimised by discouraging the use of swing moorings and placing moorings only within existing sand patches. ACKNOWLEDGEMENTS This work was funded by the Environmental Protection Authority, Western Australia. We gratefully acknowledge Mr. K. Grey who assisted with the fieldwork at Rottnest Island, Mr. A. Kirk of the Rottnest Island Board provided details of the Rottnest mooring numbers and locations. REFERENCES Bulthuis, D.A., 1983. Effects of in situ light reduction on density and growth of the seagrass Heterozostera tasmanica (Martens ex Aschers.) den Hartog in Western Port, Victoria, Australia. J. Exp. Mar. Biol. Ecol., 6~: 91-103. Cambridge, M.L. and McComb, A.J., 1984. The loss of seagrasses in Cockburn Sound, Western Australia. I. The time course and magnitude of seagrass decline in relation to industrial development. Aquat. Bot., 20: 229-243. Den Hartog, C., 1987. "Wasting disease" and other dynamic phenomena in Zostera beds. Aquat. Bot., 27: 3-14. Kemp, W.M., Boynton, W.R., Twilley, R.R., Stevenson, J.C. and Means, J.C., 1983. The decline of submerged vascular plants in Upper Chesapeake Bay: Summary of results concerning possible causes. Mar. Technol. Soc. J., 17 (2): 78-89.
EFFECT OF MOORINGS ON SEAGRASS BEDS IN AUSTRALIA 77 Kirkman, H., 1985. Community structure in seagrasses in southern Western Australia. Aquat. Bot., 21: 363-375. Kirkman, H. and Parker, R.R., 1978. Seagrasses. CSIRO Aust. Div. Fish. Oceanogr. Rep., 1974-77. Kirkman, H. and Walker, D.I., 1989. Western Australian Seagrass. In: A.W.D. Larkum, A.J. McComb and S.A. Shepherd (Editors), Seagrasses: A Treatise on the Biology of Seagrasses with Special Reference to the Australian Region. Elsevier/North Holland, Amsterdam, pp. 157-181. Larkum, A.W.D., 1976. Ecology of Botany Bay. I. Growth of Posidonia australis (Brown) Hook. f. in Botany Bay and other bays of the Sydney Basin. Aust. J. Mar. Freshwater Res.. 27:117-127. McComb, A.J., Cambridge, M.L., Kirkman, H. and Kuo, J., 1981. The biology of Australian seagrasses. In: J.S. Pate and A.J. McComb (Editors), The Biology of Australian Plants. UWA Press, Western Australia, pp. 258-293. Neverauskas, V.P., 1987. Monitoring seagrass beds around a sewage sludge outfall in South Australia. Mar. Pollut. Bull., 18: 158-164. Orth, R.J. and Moore, K.A., 1983. Chesapeake Bay. An unprecedented decline in submerged aquatic vegetation. Science, 222: 51-53. Patriquin, D.G., 1975. "Migration" of blowouts in seagrass beds at Barbados and Carriacou, West Indies, and its ecological and geological implications. Aquat. Bot., 1: 163-189.