Seagrass species

What are seagrass species?

Seagrasses are aquatic flowering plants that form meadows in near-shore brackish or marine waters in temperate and tropical regions. Australia has the highest diversity of seagrasses in the world [1], comprising more than half of the world's species, and all but one genus [2]. At the broadest level, seagrasses are differentiated into temperate and tropical species. Seagrass species can also differ in terms of the breadth of their distributional ranges (broad vs restricted), their reproductive strategies (e.g. rapid seeding, seed banks and vegetative reproduction), the degree of their persistence (ephemeral vs persistent), physiology (e.g. growth dynamics, nutrient cycling and response to disturbance) and in their ecological interactions (e.g. influence of grazing, leaf canopy structure, detritus production and epiphyte production) [3]. Assemblages of seagrass species give rise to a series of dynamic and temporally and spatially variable seagrass meadows [4]. Changes in the species composition of seagrass meadows may indicate slow but important changes in the environment, and are a suggested indicator for State of the Environment reporting [6].

Figure of the Generic Seagrass Model

Figure 1. Generic Seagrass model which categorises seagrasses on the basis of growth forms (reproduced with permission of the Marine Botany, Centre for Marine Studies, University of Queensland).

Back to top of page

What causes the species composition of seagrass meadows to change?

In broad terms, Australian seagrasses can be classified into four habitat types: river estuaries; coastal; reef and deep water [4]. The different habitat types have different ecological processes, respond to different threats and have different management requirements [4].

  • River estuary habitats can be intertidal or subtidal, and are characterised by episodic terrigenous runoff with pulses of nutrients, turbidity and reduced salinity [4]. The species composition of river estuary seagrass meadows may change in response to variable resilience of the different species to burial, anoxia and to light reduction caused by suspended sediment and eutrophication [4].
  • Coastal seagrass habitats support high levels of primary productivity and are the most biodiverse of all the seagrass habitats [4]. The primary controls on the species composition of these habitats are tidal range (intertidal or subtidal), physical disturbance caused by storms and cyclone related swell and waves, sediment movement, and the extent of grazing by macroherbivores (e.g. turtles and dugongs)[4].
  • The seagrass communities of reefs support a high level of biodiversity and are highly productive[4]. The species composition of these communities reflect low nutrient availability, unstable sediment and fluctuating water temperature and salinity [4].
  • Deep water seagrasses (15 - 58m) often form monospecific meadows [4]. The primary control on deep water seagrass meadows are low light levels (and changes in spectral composition) caused by the refraction and absorption of light in the water column and pulsed turbidity events [4].

Back to top of page

Significance of seagrass species composition

Shifts in species composition are often more sensitive indicators of ecosystem perturbation than changes in ecosystem function (e.g. production, decomposition and nutrient cycling) [5], and changes in seagrass species may occur prior to the loss of critical seagrass habitat. Shifts in dominant species of seagrass meadows may also translate into differences in benthic-pelagic coupling and the quality of habitat for fish and other organisms.

Back to top of page

Considerations for measurement and interpretation

The Department of Environment and Water Resources provides guidelines for State of the Environment reporting [6]. Seagrass floral composition should be measured at reference sites by annual assessments [6]. Methods include quadrats, line transects and leaf-blade scales [6]. Changes in seagrass species should be expressed as an estimate of change.

With recent technological improvements, remote sensing has become a cost-effective tool for monitoring and mapping the diversity, distribution and abundance of aquatic vegetation such as seagrasses, at a range of spatial and temporal scales. The principle of remote sensing of seagrasses and their environment is based on a remote sensor being able to 'see' the substrate and the vegetation growing on or in that substrate. In coastal waters, spectral scattering and absorption by phytoplankton, suspended organic and inorganic matter and dissolved organic substances restrict the light passing to and reflected up from the benthos [7]. Therefore, the spectral discrimination between aquatic plant species must concentrate on pigment-related spectral features within the visible wavelength where the light penetrates the water column and can be reflected back to the sensor [8].

There a variety of examples (see Case Studies section below) where passive remote sensing have shown to be successful (e.g. Change detection of Seagrasses in Wallis Lake and Seagrasss species, biomass and % cover maps in Moreton Bay) but it there are still areas in especially turbid or deep waters where integration of field and image data is needed to create seagrass maps (e.g. Integrated field and remote sensing approach to map seagrass cover in Moreton Bay).

A Coastal Remote Sensing Toolkit (described in more detail below) has been created to assist managers, scientists and technicians working in coastal marine environments to work out how images collected from satellites and aircraft can be used to map and monitor changes to indicators of coastal ecosystem health.

Back to top of page

Case Studies

Detection of Changes in Seagrasses Composition in Wallis Lake, New South Wales

Moderate spatial resolution multi-spectral satellite imagery was used to monitor the change over 14 years in seagrass communities within Wallis Lake, NSW (Figure 2).

Considering the first field monitoring of seagrasses only started in 1996, the use of archival Landsat data is the only way to retrospectively establish an environmental baseline in 1988 (objectively and repeatable). The Wallis Lake project has been based on archived Landsat data, however data with higher spatial resolution, and increased signal to noise ratio are currently available [9]. Although there would be a significant increase in the cost of data acquisition, hyperspectral data provide many more opportunities than multispectral imagery. Hyperspectral data have been used to successfully map rock platform vegetation, seagrass species, mangroves, salt flats and water quality parameters such as total suspended sediment (TSS), chlorophyll and coloured dissolved organic matter (CDOM) concentrations [7,10].

Figure of benthic substrate classification in Wallis Lake, NSW

Figure 2. Benthic substrate classification of the Landsat 7 satellite image collected on the 12th September 2002 in Wallis Lake [9]. The lake's benthic material was classified into sets of spectral classes representing the patterns and texture of the ecosystem. Change detection analysis also showed where the submerged vegetation community has undergone significant change from 1988 until 2002 [9].

The use of passive-optical (sensors using sunlight) remote sensing to map seagrass properties in coastal environments with constantly turbid waters is only possible in inter-tidal areas at low tides. In regions with variable water clarity, these approaches should be used on images acquired at times of clear water or low tide.

Seagrass species, biomass and %cover maps in Moreton Bay

The accuracy of high spatial resolution multi-spectral satellite imagery (Quickbird) for mapping several seagrass properties (Cover, Species and Biomass) was assessed for the Eastern Banks in Moreton Bay, South East Queensland [11] (Figure 3).

The cover and species maps were created through combination of detailed field data based on georeferenced photo transects collected in July 2004 and Quickbird imagery acquired 17th September 2004 [12, 13]. Field data were used to create training sites for supervised classification of the imagery.

Cores through the seagrass beds were collected in July 2004 and these data were used with corresponding Quickbird green band reflectance values to develop a regression formula for estimating above ground seagrass biomass for each seagrass pixel in the image.

Figure of benthic substrate classification in Wallis Lake, NSW

Figure 3. Seagrass parameter maps derived from the Quickbird satellite image collected on the 17th September 2004 above the Seagrass beds in the Eastern Banks in Moreton Bay. The maps are an example on how seagrass cover, species and biomass information can be gathered through appropriate processing of high spatial resolution, multi-spectral imagery [11].

Back to top of page

Existing information and data

Universities, museums and herbariums in each state have information on seagrasses derived from local studies. There are no on-going monitoring studies. A strategic review and development of R&D plan for Australian seagrasses was commissioned by the Fisheries Research and Development Corporation (FRDC) and follows from The Fisheries Habitat Review. The book "Seagrass in Australia: Strategic Review and Development of an R&D Plan" is available at CSIRO publishing. A document summarising this research is available at this FRDCweb site. Community-based seagrass monitoring programs, for example Seagrass-Watch, exist in some states.

In the recent edition (2007) of Seagrasses: Biology, Ecology and Conservation (Larkum, Anthony W.D.; Orth, Robert J.; Duarte, Carlos M. (Eds.)), experts in 26 areas of seagrass biology present their work in sections that relate to taxonomy, anatomy, reproduction, ecology, physiology, fisheries, management, conservation and landscape ecology. There is also a detailed chapter on the remote sensing of seagrass ecosystems.

More information on habitat removal/disturbance.

Back to top of page

Coastal Remote Sensing Toolkit

Maps of the coastal environment are needed for inventory, monitoring and management. Remote sensing instruments on various platforms (boat, aircraft and satellite) and processing techniques can be used to collect appropriate image data and convert them into maps of relevant properties. Remote sensing provides an opportunity to produce maps that can be: accurate, reliable and cost effective.There is a continually expanding array of boat, airborne and satellite based image types and processing approaches to choose from for mapping and monitoring coastal environments. A web based instructional toolkit has been created to inform managers, scientists and technicians working in coastal marine environments how images collected from satellites and aircraft can be used to map and monitor changes to indicators of coastal ecosystem health. http://www.gpa.uq.edu.au/CRSSIS/tools/rstoolkit/ [14]

Back to top of page

References

  1. Kuo, J. and McComb, A.J. 1989. Seagrass taxonomy , structure and development. In: Larkum, A.W.D, McComb, A.J. and Shepherd, S.A. (eds). Biology of seagrasses: a treatise on the biology of seagrasses with special reference to the Australian region. pp. 6-69, Elsevier, Amsterdam.
  2. Butler, A.J. 1999. Seagrass in Australia: Strategic Review and Development of an R&D Plan, FRDC Project 98/223. www.publish.csiro.au/books/bookpage.cfm?PID=2189&TXT=TOC&SITEKEY=main
  3. Dennsion, B. 2001. Moreton Bay Seagrasses, University of Queensland, Marine Botany Brochure.
  4. Carruthers, T.J.B., Dennison, W.C., Longstaff, B., Waycott, M., McKenzie, L.J., Lee Long, W.J. in prep. Seagrass habitats of north east Australia: models of key processes and controls.In Prep. and references therein
  5. Schindler, D.W. 1987. Detecting ecosytem response to anthropogenic stress. Canadian Journal of Fisheries and Aquatic Sciences 44, 6-25.
  6. Ward, T., Butler, E. and Hill, B. 1998. Environmental Indicators for National State of the Environment Reporting, Estuaries and the Sea, Commonwealth of Australia, pp. 81. www.ea.gov.au/soe/coasts/estuaries-ind.html.
  7. Dekker, A.G., Brando, V.E., Anstee, J.M., Pinnel, N., Kutser, T., Hoogenboom, H.J., Pasterkamp, R., Peters, S.W.M., Vos, R.J., Olbert, C. and Malthus, T.J. 2001, Imaging spectrometry of water In Imaging Spectrometry: Basic principles and prospective applications, vol. IV, Remote Sensing and Digital Image Processing. Kluwer Academic Publishers, pp. 307-359.
  8. Fyfe, S.K. and Dekker, A. 2001, Seagrass species: are they spectrally distinct. IEEE2001 International Geoscience and Remote Sensing Symposium, CD-ROM. IEEE, Sydney, Australia.
  9. Dekker, A. G., Anstee, J.M. and Brando, V.E. 2003, Seagrass change assessment using satellite data for Wallis Lake, A consultancy report to the Great Lakes Council and Department of Land and Water Conservation, Technical report 13/03, CSIRO Land and Water, Canberra, April 2003. (a pdf version can be found at www.clw.csiro.au/publications/technical2003/)
  10. Brando, V.E. and Dekker, A.G. (in press) Satellite hyperspectral remote sensing for estimating estuarine and coastal water quality, IEEE Trans. Geosci. Remote Sensing, (in press).
  11. Phinn, S., Roelfsema, C., Dekker, A. , Brando, V., Anstee, J.and Daniel, P. (2006) Remote Sensing for Coastal Ecosystem Indicators Assessment & Monitoring (SR). SR 30.1 Final Report: Maps, techniques and error assessment for seagrass benthic habitat in Moreton Bay. Coastal CRC Technical Report No 76, 115 pp. CRC for Coastal Zone, Estuary and Waterway Management, Brisbane. http://www.coastal.crc.org.au/pdf/TechnicalReports/76_Moreton_Bay_remote_sensing.pdf
  12. Roelfsema, C.M., Joyce, K. E., Phinn, S.R.,(2006) Evaluation of Benthic Survey Techniques for Validating Remotely Sensed Images of Coral Reefs. Proceedings 10th International Coral Reef Symposium Okinawa.
  13. Roelfsema C.M. , S.Phinn &.K.Joyce, (2006) A Manual for Using GPS Referenced Digital Photo Transects to Validate Benthic Cover Maps, Centre for Remote Sensing and Spatial Information Science. University of Queensland, Australia.
  14. (2006) Coastal Remote Sensing Toolkit. Coastal CRC and Centre for Remote Sensing and Spatial Information Science, The University of Queensland. http://www.gpa.uq.edu.au/CRSSIS/tools/rstoolkit/

Contributors

Janet Anstee, CSIRO Land and Water
Arnold Dekker, CSIRO Land and Water
Stuart Phinn, UQ, Centre for Remote Sensing and Spatial Information Science
Chris Roelfsema, UQ, Centre for Remote Sensing and Spatial Information Science

Back to top of page