Landing Site Suitability Models

Introduction: We have explored the use of GIS suitability models for the screening and analyzing of potential sites for Mars landers using engineering and scientific constraints in an iterative manner. While initially performed on locally held data sets, these analyses have now been extended to use widely distributed data from standardized Web Mapping Servers (WMS) [1]. The ability to exploit a wide variety of data within a single analysis tool offers the opportunity for community wide participation in landing site selection. We have created several preliminary models within the Geographic Information System (GIS) software package ArcMap by the Environmental Systems Research Institute (ESRI). We have extended the capability of ArcMap to use Internet streamed data sets within the models as well as hosting a user-controlled model entirely from a web interface. Motivation: Selection of landing sites for martian rovers can be a very tricky balance between landing safely and choosing scientifically interesting areas. Spacecraft safety is always the primary concern, but within all of the possible " safe " sites, those that are of greater scientific interest are the ones to be considered. Bringing in a wealth of available data sets to help with this choice requires that all data are co-registered and in the same coordinate and projection system. GIS technology provides a common substrate for the analysis of these disparate data sets, and for the application of both engineering and scientific constraints to select out the most promising and suitable areas. Method: For the purposes of testing the GIS approach, we chose several of the engineering constraints used in the selection of landing sites for the Mars Exploration Rovers as described in Golombek et al [3]. Our simplest model includes MOLA topography [4] (to access elevation and slope), MOLA roughness, and TES thermal inertia [5], but we have also tested other parameters like proximity to hematite (Figure 1). A sample model run reclassifies each layer and ascertains landing site suitability based on weighted results for each constraint. The weights for each layer are user-definable and interactive which allows the user to easily test many different scenarios. An example of an absolute constraint would be using the MOLA topography to determine if the elevation were low enough, which is necessary to allow the descent vehicle to slow down. An example parameter that could be weighted from " optimal " to " questionable " to " unacceptable " would be …