Mapping shellfish distribution using hyperspectral remote sensing

A majority of all baseline shellfish maps are currently produced through ground surveys and manual aerial photo interpretation, a tedious process which is time-consuming and prone to human errors. The first objective of this research was to investigate and document the feasibility of using remotely sensed imagery to identify and classify intertidal shellfish resources. A specific task was to spectrally identify the differences between mud and shellfish using a combination of hyperspectral remote sensing data and extensive in situ spectral data. A second objective was the identification of spectrally relevant portions of the electromagnetic spectrum that are useful in identification of shellfish and to determine if there is a statistical difference between mud and shellfish.;This research found that it was feasible to accurately identify shellfish and distinguish shellfish from mud. It was found that with the incorporation of in situ derived spectral endmembers, the accuracy of the mapping procedure was lower than deriving spectral endmembers directly from the remotely sensed imagery. It was also noted that changes within the habitat could have occurred between the acquisition of the remotely sensed HyMAP and AISA imagery and the acquisition of the in situ data may have contributed to lower accuracy results. Secondary objective results indicate that the visible and near infra-red portions of the electromagnetic spectrum are the most useful for discriminating between shellfish and mud. While statistical analysis showed there was a significant difference between mud and shellfish, there was also a high degree of statistical differences within the collected shellfish data that a degree of uncertainty remains.;The use of remotely sensed imagery, in situ ancillary data and field verification does result in shellfish maps that are less error prone than previous methodologies and allows for the mapping of remote or hard to access areas of shellfish resources.;It is hoped that the creation of a repeatable, timely, and cost effective mapping technique that is less prone to error for the creation of baseline shellfish maps will benefit the coastal community at national and international levels.