Determining the enhanced separability of seagrass reflectance spectra based on identifying key markers and environmental factors: Case Study of Gulf St Vincent South Australia

Abstract

Seagrasses are a vulnerable and declining coastal habitat which are crucial to providing shelter and substrata for aquatic microbiota mollusks invertebrates fishes and sea turtles More accurate mapping of seagrasses is imperative for their survival as a sustainable natural resource but is encumbered by the lack of data and data processing techniques for reflectance spectra representing the optical signatures of individual species Moreover before reflectance spectra could properly be used as remote sensing endmembers factors that may obscure the detection of reflectance signals must also be assessed Objectives of this study are seven-fold: (1) to determine distinguishing characteristics of spectral profiles for sand versus three temperate seagrasses (Posidonia Amphibolis and Heterozostera); (2) to evaluate the most efficient derivative analysis method of spectral reflectance profiles for discriminating between benthic types; and to assess the influences of (3) site location; (4) filtered marine water; (5) epiphytes; (6) natural growing depth; and (7) particular seagrass genera on spectral response signals based on varying degrees of epiphyte presence and natural growing depth Results show that the 566:689 and 566:600 bandwidth ratios are useful for differentiating seagrasses from sand and from detritus and algae respectively First-derivative deconvolution reflectance spectra in general is the most efficient derivative-based method for identifying unique non-contiguous bandwidths throughout the visible light spectrum particularly with deconvolution analyses further helping to reveal and isolate 11 key wavelength dimensions (417 456 474 491 522 590 605 621 631 649 and 681 nm) Both differences between sampling sites as well as marine water that is filtered and unfiltered show no statistically significant effect on reflectance endmembers These results can potentially be applied when extrapolated to other locations; albeit regarding these two factors first caution should be used due to proximity of sampling sites and second it is possible that water quality at the time of data collection was probably adequately sufficient and therefore negligible in differences Epiphytes significantly dampen bottom-type reflectance throughout most of the visible light spectrum excluding bandwidths at 485–510 and 645–680 nm (p < 0 005); these exclusionary bandwidths may be useful for assessment of seagrasses regardless of the influence of epiphyte presence Growing depth at which seagrasses are naturally found does influence reflectance spectral responses with the detection of deeper seagrasses (2 to

Date of Award	2019
Original language	English
Supervisor	Chih-Hua Chang (Supervisor)