An ROV toolbox for optical and acoustical seabed investigations

Abstract

This thesis describes a set of optical and acoustical methods for documenting a research site using underwater vehicles. The need for observations in cold-water coral research has formed the design problem in this work. A common survey protocol template is developed to describe video-, bathymetry-, photo mosaic- and photogrammetry surveys. For photogrammetry, the body of knowledge is small, and it was necessary to develop a new method rather than adapting existing methods. Axiomatic design methodology is applied to create the protocol template. Functional requirements for documenting these biological communities have been established and matched with solutions found in academic literature, industry standards and established best practice from hydrography, offshore subsea sector and related research disciplines.

The protocols provide guidelines for design of all phases of scientific surveys from planning to survey reporting. The first step in the survey design is to evaluate the practical constraints, the purpose of the operation and to establish requirements for the desired data products. Step two is to choose and configure the survey instruments using the information from step one. To create a plan for the operation containing a description of central parameters like survey lines, line spacing and necessary vehicle velocity the results from step one and two are applied. Step four is the final step in the survey design process where the data processing pipeline is compiled based on the previous design steps. The survey protocols presented are followed by data examples collected by the candidate to show their relevance and application.

The survey methods presented are complimentary with different levels of coverage and data resolution to cover a range of scales to address subjects like target confirmation, search, coral reef mapping and detailed documentation of individual specimen. Sonar surveys represent high coverage and low data resolution, while photogrammetry on the contrary provides high data resolution and low coverage.

This thesis may contribute to the establishment of a common practise for marine science using ROV for data acquisition. This can enhance the value of ROV-based surveys. The efficiency in data acquisition operations will increase and the value of the collected data increases if one succeeds standardizing operations.