Abstract

The spatial distribution of volcanic activity has been the subject of scientific enquiry for more than 350 years. Nevertheless, it has been only until recently that modern techniques of analysis started to be used to characterize such distribution. As a result, in the past 40 years many methods were devised, or adopted from other fields of study, aiming to achieve that purpose. Ironically, the plethora of now available methodologies has made problematic the selection of one method of study to characterize the spatial distribution of volcanic vents. In addition, given the diversity of methods, and their underlying assumptions, there are issues concerning the form in which the results of any specific method should be interpreted in a volcanic context. In this work I make emphasis on the relationship between the embedded assumptions of several common methods of study with the concept of a “natural partition”. Throughout this work it is remarked that 1) it is extremely important to gain conscience about the several assumptions implicit on each method of analysis, 2) never loose sight that not all volcanic systems are equal, and 3) that it is a myth that any quantitative method can provide reliable information about any system. The convenience of assessing more than one conceptual model to explain the formation of a cluster-subcluster structure is also mentioned. Overall, it is shown that the complexity of volcanic phenomena cannot be encapsulated by using a single method of study, or an automatized selection of parameters. Thus, given the complexity and diversity of volcanic phenomena, there is no particular method, nor set of methods, that can be recommended to be used on every situation of interest. Yet, as a golden rule, it is suggested that more than one method of analysis is used at every location of study, looking for congruency of results within a range of spatial scales.

Full Text
Published version (Free)

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