Using spatial data support for reducing uncertainty in geospatial applications

Abstract

Widespread use of GPS devices and ubiquity of remotely sensed geospatial images along with cheap storage devices have resulted in vast amounts of digital data. More recently, with the advent of wireless technology, a large number of sensor networks have been deployed to monitor many human, biological and natural processes. This poses a challenge in many data rich application domains now: how to best choose the datasets to solve specific problems? In particular, some of the datasets may be redundant and their inclusion in analysis may not only be time consuming, but also lead to erroneous conclusions. On the other hand, excluding some of the datasets hastily might skew the observations drawn. We propose the concept of data support as the basis for efficient, cost-effective and intelligent use of geospatial data in order to reduce uncertainty in the analysis and consequently in the results. Data support is defined as the process of determining the information utility of a data source to help decide which one to include or exclude to improve cost-effectiveness in existing data analysis. In this paper we use mutual information--a concept popular in information theory as a measure to compute information gain or loss between two datasets--as the basis of computing data support. The flexibility and effectiveness of the approach are demonstrated using an application in the hydrological analysis domain, specifically, watersheds in the state of Nebraska.