Jeffrey L. Star

Requirements and principles for the implementation and construction of large-scale geographic information systems

Abstract This paper provides a brief survey of the history, structure and functions of ‘traditional’ geographic information systems (GIS), and then suggests a set of requirements that large-scale GIS should satisfy, together with a set of principles for their satisfaction. These principles, which include the systematic application of techniques from several sub-fields of computer science to the design and implementation of GIS and the integration of techniques from computer vision and image processing into standard GIS technology, are discussed in some detail. In particular, the paper provides a detailed discussion of questions relating to appropriate data models, data structures and computational procedures for the efficient storage, retrieval and analysis of spatially-indexed data.

Oceanographic biooptical profiling system

A new oceanographic instrument to measure underwater optical, biological, and physical properties has been designed, built, and used extensively at sea. The new instrument system is a significant advance that permits optimum sampling strategies using ship, aircraft, and satellite optical sensors for ocean research. The design criteria for the biooptical profiling system included the rapid acquisition of data to accommodate shipboard synoptic sampling; the measurement of the necessary parameters for providing appropriate contemporaneous surface data for remote sensors; the compatibility of the data acquisition system with conventional conductivity temperature depth-type cables and winches on the oceanographic fleet; the capability of real-time display of data along with rapid preliminary data reduction at sea. The primary instrument package in this profiling system is a new microprocessor controlled multiwavelength spectroradiometer. A description of the instrument and examples of data are presented.

Performance analysis of image processing algorithms for classification of natural vegetation in the mountains of Southern California

Abstract The Earths forests fix carbon from the atmosphere during photosynthesis. Scientists are concerned that massive forest removals may promote an increase in atmospheric carbon dioxide, with possible global warming and related environmental effects. Space-based remote sensing may enable the production of accurate world forest maps needed to examine this concern objectively. To test the limits of remote sensing for large-area forest mapping, we use LANDSAT data acquired over a site in the forested mountains of southern California to examine the relative capacities of a variety of popular image processing algorithms to discriminate different forest types. Results indicate that certain algorithms are best suited to forest classification. Differences in performance between the algorithms tested appear related to variations in their sensitivities to spectral variations caused by background reflectance, differential illumination, and spatial pattern by species. Results emphasize the complexity between the...

Natural oil seep detection in the Santa Barbara Channel, California, with Shuttle Imaging Radar

Natural submarine oil seeps in the Santa Barbara Channel, California, were detected by the first Shuttle Imaging Radar (SIR-A). Oil slicks on the ocean are seen in radar imagery as areas of decreased radar signal return that result from a damping of surface roughness. Orbital radar imagery shows promise as an effective and efficient means of mapping submarine oil seeps as sources of naturally occurring oil slicks.

A comparison of maximum likelihood‐based supervised classification strategies

Abstract Supervised land cover classification strategies for Thematic Mapper image data are tested in two southern California study sites. Effects of training field aggregation, prior probabilities, and nonnormality in spectral data are investigated. Findings are summarized using both PCC and Kappa statistics. Error matrices are presented in graphic form to enhance comparison among images and methods. Differences in overall map accuracy between tested strategies are generally significant, and certain strategies tend to produce consistent patterns in the error matrices of all image products. Results of these supervised classifications are also briefly compared to previously published results for unsupervised techniques using identical image data products.

MAPWD: an interactive mapping tool for accessing geo-referenced data sets

Abstract Many scientists use geographic location as a fundamental attribute to identify interesting earth-science data sets. The specification of this attribute in database queries is both tedious and prone to error. MAPWD is a graphical tool which simplifies this specification by allowing the user to point to areas of interest using an electronic map, instead of manually entering coordinate values. With this tool the user is able to identify regions of interest using maps with comparable detail at a wide range of spatial scales. This functionality has been achieved by hierarchically structuring the database of map vector data into separate files based on both geographic location and resolution.

Support for global science - Remote sensing's challenge

Abstract Remote sensing today uses a wide variety of techniques and methods. Resulting data are analyzed by man and machine, using both analog and digital technology. The newest and most important initiatives in the U.S. civilian space program currently revolve around the Space Station complex, which includes the core station as well as co‐orbiting and polar satellite platforms. This proposed suite of platforms and support systems offers a unique potential for facilitating long term, multi‐disciplinary scientific investigations on a truly global scale. Unlike previous generations of satellites, designed for relatively limited constituencies (e.g., Landsat for the land scientist and Seasat for the oceanographic community), Space Station offers the potential to provide an integrated source of information which recognizes the scientific interest in investigating the dynamic coupling between the oceans, land surface, and atmosphere. Earth scientists already face problems that are truly global in extent. Probl...

Sources Of Error In Thematic Classification Of Remotely Sensed Imagery

From a statisticians point of view, the input datasets are rarely examined to determine their underlying frequency distribution; it is just assumed that the data are normal enough, and that the deviations from normality are unimportant. It is not clear how deviations from a hypothetical multivariate normal might affect the power of the classification process, and there is ample evidence in the literature that, at a minimum, the spectral channels are correlated. From a practioners point of view, in a supervised classification the number of training fields for developing a statistical description of a given class is usually arbitrary. It is unclear how small changes in the details of the training field selection process affect the quality of the derived thematic information. The start of an examination of this latter problem is discussed.