Abstract
Visualization of multidimensional data helps in understanding complex systems and environments. We present here a red, green, blue (RGB) visualization method that can serve to display environmental properties. The saturation of each color is used to represent the concentration of a given property. The implementation of that figure is illustrated through visualization of three dissolved inorganic nutrient concentrations along a vertical transect of the Mediterranean, as well as through a vertical time series of three phytoplankton group cell numbers. The RGB figures show well known properties of the water column. In addition, they reveal some lesser-known properties, such as regions in shallow water in which the ratio of phosphorus and silica to nitrogen is high, and a deep eukariotic phytoplankton community. Visualization of such data is usually performed with three separate contour or surface plots, and occasionally two properties are presented as an overlay in a single figure. The RGB figure offers a better way to visualize the interactions among the three separate plots than is commonly available.
Highlights
Classification and representation of multidimensional data are of great scientific interest
An increase in the proportion of Pro in summer is seen, in the form of bluer color on the right side of Figure 4 at a depth of ca. 80 m. Another feature revealed by the RGB representation is the existence of a eukaryotic phytoplankton (Euk) dominated phytoplankton community at depth of ca. 200 m throughout the period, which appears as a redder stripe at this depth
We presented a novel methodology for visualizing data containing multi-dimensional dependent variables in a single plot, facilitating interpretation of the data
Summary
Classification and representation of multidimensional data are of great scientific interest. Methods such as the principal component analysis and k-means classify the data while losing the original data values. A visualization method of four dimensional data in two dimensional space was developed, presenting data values by color (hue) and uncertainty by saturation. Roederer et al [2] and Saadatinejad et al [3] addressed the problem of displaying multidimensional data in two dimensional space by using the color matrices of the RGB model, showing raw flow cytometry results and seismic data through color saturation, respectively. Following growth in the volume of data [4] as well as the increase in data complexity, new multidimensional data visualization methods must be developed to facilitate our understanding of complex systems
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
