Scientific workflows for the geosciences: An emerging approach to building integrated data analysis systems

Abstract

Scientific method and the influence of technology Due to the increasing number and sophistication of data acquisition technologies, the amount of raw data acquired has vastly increased over the last couple of decades (Berman, 2008). This explosion of scientific data, growth in scientific knowledge, and the increase in the number of studies that require access to knowledge from multiple scientific disciplines amplify the complexity of scientific problems. In order to answer these “grand challenge” scientific questions, scientists use computational methods that are evolving almost daily. The basic scientific method, however, remains the same for the individual scientist. Scientists still start with a set of questions, then observe phenomena, gather data, develop hypotheses, perform tests, negate or modify hypotheses, reiterate the process with various data, and finally come up with a new set of questions, theories, or laws (http://en.wikipedia.org/wiki/Scientific_method). A recent change in this scientific method is that it is continuously being transformed with the advances in computer science and technology. The simplest examples of this transformation are use of personal computers to record scientific activity and the way scientists publish and search for publications online. More advanced technologies within the scientific process include sensor-based observatories to collect data in real time, supercomputers to run simulations, domain-specific data archives that give access to heterogeneous data, and online interfaces to distribute computational experiments and monitor resources.