Abstract
The large kinetic energy inherent to eddies is a strong modulator of the global climate, ocean circulation, productivity, and freshwater transport. The importance of creating a census of eddies around the ocean has been a pursuit for oceanographers around the world since satellites and measurement instruments have been publicly available. GIScience plays an important role in this pursuit thanks to its ability to transform continuous data to discrete features with specific attributes. This paper deals with the limitations of GIS resources such as resolution and methodologies that make it possible to create a snapshot of these dynamic structures in the upper levels of the ocean. A new approach is incorporated trough the use of AI algorithms that aided the author to understand the theory behind a set of twelve papers that adress the constantly growing fields overlapping oceanography and geographic information science.
Keywords: geographic information science, eddy detection, ocean currents
Introduction
The attention that in recent years has been given to the use of GIS and its innate ability to transform continuous surfaces into valuable discrete attributes means that the surface of the ocean can be classified, categorized and hence, understood, at least in a shallow way. As a part of the meridional overturning circulation system, eddies are categorized as dynamic structures that break out of the main current, becoming a fractal part of the structure. As currents, they too serve as regulators of temperature, salinity and sea level but in a smaller scale. Geographic information science proved to be a powerful tool to oceanographers helping them discriminate these structures from satellite imagery and continuous datasets on a methodology that roughly sketched involved identifying coherent vortices out of sea surface elevation rasters trough the assessment of geostropic relations. The objective of this paper is to address the heterogeneous approaches that the scientific community has taken regarding the detection of eddies in different parts of the ocean. One of the main takeaways of this process is the different way that this problem has been resolved but also to track the decision processes that scientist have selected considering the availability of resources, the study area variability, the initial input (height or temperature) and the approach taken at understanding the nature of fluids in the ocean. In order to give a panoramic view of this area of oceanography twelve scientific papers have been selected to be compared and computed by an artificial intelligence algorithm. These documents became a chatBot that provided the author with answers fed exclusively with data from this ingested input. In some instances the intelligence served as a way to understand processes and theoretical concepts, whilst in other cases became a way to retrieve an specific attribute that manifested in different ways in the literature. This part of the analysis is further explored in the methodology section where a subset of 25 questions was analyzed to assess the effectiveness and the performance of the machine learning algorithm to retrieve data from all of the sources.
In order to have a good sample database SCOPUS database was used, and even tough at first, the objective of this paper was to understand the different representations of oceanic data in cartographic outcomes, a secondary approach was selected to minimize the extent of the analysis and solely focus on the detection of ocean eddies using continuous surfaces. This decision was made considering resources such as time and prior knowledge of the topic.
This complexity reduction allowed the author to confidently understand the theoretical basis of current dynamics in oceanic databases and their interpretation in GIS environments, sketch the process in simple terms and build connections and comparisons in the formal aspects of how each paper approach the same idea.
The PRISMA methodology was used in scopus to access the best possible papers as input. The overall theme of the selected bibliography was representation of oceanic parameters in geographic information science with a special emphasis in ocean eddies.
