Martian Dust Storm as imaged on November 18, 2012
With all the interesting images and news from Curiosity we almost would forget what other Mars probes we have at Mars. The Mars Reconnaissance Orbiter (MRO) is one of these Martian assets that are at our disposal. The image above shows observations made by the Mars Color Imager on MRO being stitched together into a nearly global mosaic of a dust storm on Mars. These observations were made on 18 November 2012. On this map the locations of both rovers, Opportunity and Curiosity are labeled. Small white arrows outline the area where dust from the storm is apparent in the atmosphere.
To acquire this image the MRO had to perform high angle roll manoeuvres which leads to some loss of data which leads to the black areas in the mosaic image. This combined image also suffers from some blurry areas running from bottom to top of the image. This blurriness is the result of high ‘off-nadir’ viewing position of the low orbit of the satellite while taking this set of images.
The Mars Color Imager is operated by Malin Space Science Systems, San Diego, California.
important satellite and aerial imagery parameter, off-nadir angle (ONA). ONA defines the angle that the camera (both satellite and aerial) lens looks at the earth. At-nadir is the point the camera is looking directly down at the ground; while the off-nadir angle would tell you how far from this point (in degrees) the lens is tilted.
Off-nadir further explained
In the picture to the left the camera lens is portrayed with a black diamond. Imagery collected at-nadir would be directly below the black arrow. The red line represents imagery collected off-nadir; and the off-nadir angle would be the angle between the black and red lines (in degrees).
ONA can impact the resolution, building lien and accuracy of both aerial and satellite imagery.
First we will address its impact on resolution. Cameras on satellites also have a limited number of sensors. When a satellite images the ground right below them (which actually never happens), the amount of ground collected in a single image is minimized so each sensor will cover the minimum amount of ground possible for that satellite, for example 41-cm for panchromatic data on the Earth-satellite GeoEye-1. As the camera tilts to capture imagery, the amount of ground imaged increases as does the ONA; the result is lower resolution in the imagery as the sensors are limited on the satellite but the amount of ground to be imaged is now larger. At 30 degrees ONA, GeoEye-1 collects imagery with 50-cm resolution where as the Earth-satellite WorldView-2 would collect imagery with ~56-cm resolution. The impact of ONA on the clarity/resolution of imagery is noticable and needs to be corrected to make the image as speaking as we want it.
Off-nadir angle also controls a phenomenon known as building lien. When imagery is collected directly at nadir, only the tops of objects in the imagery would be visible. As the off-nadir angle increase, so too would the tilt of the objects in the imagery thus their tops and increasingly their sides would be visible. Imagery collected at 45 degrees ONA is often referred to as oblique imagery. When objects lien, they can lien in different directions. When aerial imagery is collected, there is a small amount of data right below the camera that is directly at nadir. As you move in either direction from this point, the ONA increases and so does building lien. Objects will actually lien in opposite directions as you move away from this point as well. When satellite imagery is collected, the optics are extremely consistent across the large focal plane that is offered with this technology. In fact, there is typically less than a 2 degree variation in ONA from edge to edge of a satellite imagery tile which can exceed 14-km wide, for example, with the Earth-satellite QuickBird imagery. As a comparison, an aerial imagery tile is typically about 2-km wide once it is stitched in with surrounding collections.
High off-nadir angles can impact the accuracy of imagery products. As the ONA of imagery increases, there is also an increase in the horizontal displacement of each pixel caused the parallax effect. While collecting ground control and then orthorectifying will help to reverse the horizontal displacement, as the horizontal displacement increases so too does the importance of accurate elevation data (i.e. DEMs) in the process. Many times people rely upon free sources of DEM data in the orthorectification process as extracting highly accurate elevation data from stereo imagery is can be expensive as compared to the mono imagery itself. As such, a high ONA in imagery will lower the accuracy of the final product if often inferior open source elevation data (e.g. NED, SRTM) is used over highly accurate elevation data.
It is clear that the magic of Michael Malin and his team is needed to give us the accurate imagery of Mars that we seek. Knowing how well Michael’s team operates around Mars we are confident that what they show us is the best we humans can do and tells a correct story of what is happening on Mars.