Sand dunes are among the most widespread aeolian features present on Mars, serving as unique indicators of the interaction between the atmosphere and surface. On a planetary body, dunes accumulate where a supply of sand-sized grains exists or may be abraded, is carried downwind by winds of saltation strength, and is subsequently deposited where these winds weaken below the threshold for sand transport. As a result, the study of dune processes contributes to both atmospheric and sedimentary science. Both the presence and morphology of sand dunes are sensitive to subtle shifts in wind circulation patterns and wind strengths, which are thought to be influenced by changes in Martian orbital parameters. The spatial distribution of aeolian sand relates to patterns of sedimentary deposition and erosion of source materials, giving clues to the sedimentary history of the surrounding terrain. Dunes are particularly suited to comprehensive planetary studies in part because they are abundant on the Martian surface over a wide range of elevations and terrain types, and in part because they are large enough to be studied using the wide suite of spacecraft data now available. Thus a global scale study of Martian dunes serves a dual purpose in furthering the understanding of both climatic and sedimentary processes, two fundamental topics currently driving Martian science. The digital dune database makes it possible to look at dunes in a global context, comparing their geographic location and attributes to other global coverages, such as geologic maps, GCMs, MOLA and TES. Such comparisons provide significant perspective on local, regional, and global-scale aeolian processes that have shaped and continue to influence the surface of Mars. Mars Global Digital Dune Database Process: The database is based on dune forms and (or) deposits that are located using calibrated THEMIS IR images (RDRs). Each THEMIS IR image is ~ 32 km in width, and can be greater than 6000 km long, providing an areal coverage greater than 180,000 km2. For most daytime images, the sensor acquires all nine bands ranging in wavelength from 6.8-14.9 m. Nighttime IR images are usually acquired using only bands 4, 9 and 10 (bands centered on 8.56, 12.57, and 14.88 ?m, respectively). Band 9 was chosen as the default image for this study because it offers the highest signal to noise ratio (SNR) at night, high SNR during the day, and is included in every acquisition. An initial data set of THEMIS band 9 images covering orbits 816-9601 (spanning 02/2002 - 02/2004 and Ls = 0.085º-358.531º), comprising more than 30,000 images planet-wide, was chosen as the basis for the construction of the database. This provided ~98% nighttime and ~75% daytime areal coverage of Mars. Images containing dunes were identified using THV (Interactive THEMIS IR Viewer written in Research Systems Incorporated's (RSI) IDL® software at the USGS in Flagstaff (www.mars-ice.org)). Despite better nighttime coverage, approximately 75% of the images identified as containing dunes were daytime images, where dunes appear brighter than the surrounding area, indicating a warmer relative temperature. On nighttime images, where dunes are often darker than the surrounding area and show less tonal variation within a dune field, the dunes were more difficult to locate. After selection, the images were map projected using Integrated Software for Imagers and Spectrometers (ISIS) into Environmental Systems Research Incorporated (ESRI) ArcMap® GIS software. Polygons were drawn around all areas that were considered to be possible dune field candidates based on the identification of dune form (where discernable) and tonal (relative thermal) contrast with surrounding material, commonly at a scale of ~1:75,000. While medium to large dune fields could often be identified at 100m/px resolution, non-optimal image quality or small feature size sometimes precluded reliable feature identification. Questionable dune fields were verified using higher resolution THEMIS VIS or MOC NA images, where available. When review of higher resolution images revealed dunes that were too small to be seen on IR images or dunes that fell outside IR image boundaries, polygons were added to the database. We commonly truncated a digitized polygon at the boundary of an image in order to preserve the unknown nature of the region outside of the selected image's extent. Mars Digital Dune Database - Completeness of Database The database covers 65º N to 65º S. It will be expanded to cover the entire planet in later versions. Although we have attempted to include all dune fields between 65º N and 65º S, some have likely been excluded for two reasons: 1) incomplete THEMIS IR (daytime) coverage or 2) resolution of THEMIS IR coverage. The importance of the first reason, incomplete daytime coverage, became apparent during the process of building the database. When we began building the database we started with the first THEMIS orbit (816), chose an arbitrary upper limit of orbit 9601, and systematically reviewed the over 30,000 THEMIS IR images included in that range. This provided 75% daytime and 98% nighttime planet-wide coverage. Although 98% nighttime coverage sounds nearly complete, dune fields were sometimes difficult to identify on nighttime images, so actual coverage may be closer to the 75% daytime coverage figure. The second reason is the resolution (100m/px) of THEMIS IR images. It allowed us to locate only moderate to large dune fields. The smallest dune fields in the database are ~ 1 km2 in area. While the moderate to large dune fields are likely to constitute the largest compilation of sediment on the planet, smaller stores of sediment of dunes are likely to be found elsewhere via higher resolution data. Thus, it should be noted that our database excludes all small dune fields and some moderate to large dune fields as well. Therefore the absence of mapped dune fields does not mean that such dune fields do not exist and is not intended to imply a lack of saltating sand in other areas.
