Modeling the Potential Swath Coverage of Nadir and Off-Nadir Pointable Remote Sensing Satellite-Sensor Systems

Pointable sensor systems onboard many earth resources satellites today, particularly the higher spatial resolution sensors, provide for a near infinite set of collection opportunities. Satellite orbits of these systems are not systematic repetitive tracks. Predicting future collection opportunities requires predicting where the satellite will be and then computing the potential swath coverage from a pointable sensor along these orbits. While each agency or company models its own satellite-sensor systems, few publicly available sources exist for mapping future satellite ground tracks. Evaluating collection opportunities from multiple satellite-sensor systems from different agencies/companies is problematic. The purpose of the research described in this article was to develop a generic approach for modeling future satellite-sensor collection opportunities. In this article, formulae are developed for computing the potential swath coverage, and an algorithm is designed for constructing the potential swath coverage area. The solution to the swath coverage problem is based on spherical trigonometry, a well known map projection (i.e., azimuthal equidistant map projection) used in an unconventional dynamic form, and a satellite orbital propagation model. We demonstrate how the computation of the swath coverage area can be accomplished using a temporal series of re-centered map projections. Cartography and Geographic Information Science, Vol. 35, No. 3, 2008, pp. 147-156 Introduction Determination of the potential ground area imaged by a satellite-sensor system on future revisit dates was once a somewhat simple endeavor. Early earth resources satellites (e.g., Landsats 1-5 and 7, or the Defense Meteorological Satellites) carried sensor systems that only imaged in a nadir pointing fashion. Also, their orbital characteristics were such that their ground tracks were repetitive (e.g., the track sequence was repeated every 16 days). The familiar path-row maps (either hardcopy or the online web form) for the Landsat series, for example, were appropriate for such a stable, repetitive collection sequence. Such systematic orbits and nadir-only imaging allowed for elegant conic or cylindrical map projection solutions showing the ground-projected satellite tracks (Snyder 1987). Evaluation of if and when an area could be imaged by a specific satellite sensor were almost trivial. The pointable sensor systems onboard many earth resources satellites, particularly the higher spatial resolution sensors (e.g., Quickbird 2, Orbview-3), provide a near infinite set of collection opportunities. The satellite orbits of these systems are not systematic repetitive tracks. They do not need to be repetitive as the sensors can point off-nadir and thus, “revisit” geographic locations. Predicting future collection opportunities requires predicting where the satellite will be and then computing the potential swath coverage from a pointable sensor along its orbit. While each agency or company models its own satellite-sensor systems, few publicly available sources (e.g., web sites) exist for mapping the future satellite ground tracks. Evaluating collection opportunities from multiple satellite-sensor systems is problematic. Those web sites offer, for the most part, very small-scale maps of future ground tracks, making precise, future collection opportunity predictions difficult. The goal of this research was to develop a generic approach for modeling future satellite-sensor collection opportunities. The problem studied here is the computation of a polygonal shape representing an area that may be imaged by a satellite with a pointable sensor on a single overpass. We refer to this as a potential swath coverage area in a 2-dimensional form and potential swath coverage in 1-dimensional form (Figure 1). The relevant sensor angles are instantaneous field of view (IFOV), field of view (FOV), and the maximum off-nadir angle of the pointable sensor. The ground-projected distances for these angles are pixel size (for instantaneous field of view), swath width (for field of view), and potential swath coverage (for the maximum off-nadir angle of the pointable sensor). The astrodynamics and lidar communities use the term boresight angle to refer to the off-nadir angle for which the FOV is Michael E. Hodgson and Bandana Kar, Department of Geography, University of South Carolina, Columbia, South Carolina 29208. Email: [email protected]; kar@mailbox. sc.edu. Tel: 803-777-8976; Fax: 803-777-4972.