As the Arctic climate warms, permafrost is melting rapidly, causing dramatic changes in the landscape. High-resolution aerial photography is an ideal tool to analyze landscape evolution in long-term studies of permafrost thawing dynamics. The DLR Laboratory for Optical Sensor Systems has developed the Polar 18 Modular Aerial Camera System (MACS), designed for high-precision 3D reconstruction and monitoring of Arctic terrain under harsh conditions, to achieve improved image quality and ultimately to better assess the impact of climate change on permafrost landscapes.
Unlike traditional aerial photography, MACS Polar 18 collects data sets in 4-band multispectral (RGB-NIR) mosaics, 3D point clouds and digital surface models. When the ground area of interest contains both very dark areas, such as water or bare soil, and very bright surfaces, such as snow or ice, multiple integrations can be collected for each scene to avoid under- or over-exposed pixels. The Polar 18 version of MACS is specially tuned to work reliably in harsh arctic regions exposed to extremely low temperatures and harsh light conditions.
MACS sensor unit
The MACS Polar 18 sensor is equipped with three SVS-Vistek HR series GigE industrial cameras with 16 megapixel resolution: two RGB cameras with overlapping right and left views and one camera with near infrared (NIR) wavelength. The cameras are housed in a sensor array along with inertial sensors that detect acceleration, rotation, and velocity. The sensors are connected to a main computer for processing and a Global Navigation Satellite System (GNSS) receiver.
Explore the North Pole
In the summers of 2018, 2019 and 2021, scientists from the German Aerospace Center (Berlin, Germany) and the University of Potsdam (Potsdam, Germany) deployed MACS Polar 18 to capture tens of thousands of images in a series of aerial photography campaigns. Final results are expected to be published in 2024. The flights imaged 1,500 square kilometers of permafrost-affected terrain in northwestern Canada and northern and northwestern Alaska. MACS was carried by two Basler BT-67 aircraft modified for harsh environments, with the sensors placed in the aircraft’s ventral ports. The aircraft fly in a grid pattern suitable for photogrammetric processing.
The two SVS-Vistek RGB and IR cameras were set to capture images with a resolution of 4864 x 3232 pixels at a continuous rate of four frames per second and were electronically triggered so that image exposures began at exactly the same time. Camera parameters other than exposure time were fixed to minimize motion blur during flight. The cameras also sent electrical pulses to the GNSS receiver, generating information that records the exact time, position, and attitude of the capture. That data was written to the RAW image file. The cameras were only switched off during takeoff and landing, when low cloud cover appeared locally, over large bodies of water or sensitive infrastructure, during sharp turns, or when the hard drive used to store the data was running low on space.
The acquired raw Bayer pattern images are stored in the proprietary MACS format. Several software packages were used in the workflow, but most of the pre- and post-processing data processing was automated by Python scripts and WhiteboxTools. Additionally, the images are processed using the Pix4Dmapper mosaic tool to combine multiple images with overlapping areas to create a seamless image. The re-formation of each mosaic is performed automatically using elevation information calculated internally in Pix4Dmapper. The mosaics create large 3D maps that allow visual observation of permafrost thaw over the years.
Future Applications
In this and other studies, it has been demonstrated that MACS can open up new possibilities not only for data analysis, but also for the exploration and validation of different landform types. This includes monitoring coastal erosion; analyzing ice-wedge polygons; monitoring subsidence due to snowmelt and assessing potential impacts on infrastructure; detailed analysis of recent and past burn scars; detecting lake drainage systems; inspecting all shrubs and trees in scrub tundra; or quantifying beaver dams and dens.
