An improved approach of total freeboard retrieval with IceBridge Airborne Topographic Mapper (ATM) elevation and Digital Mapping System (DMS) images

Abstract

The existence and detection of leads are critical to obtain a local Sea Surface Height (SSH) reference for computing total freeboard and sea ice thickness from NASA's IceBridge Airborne Topographic Mapper (ATM) elevations. However, the shaded areas of the Digital Mapping System (DMS) images and the biased ATM elevations impact the correct determination of leads and SSH. This study develops an automated approach to overcome the above challenges and to correctly determine SSHs by combining DMS images, ATM L1B's apparent reflectivity and statistical discrimination. Dynamic pixel intensity thresholds are established to classify leads under different solar illuminations. This automated approach is then validated by manual selection in detecting SSHs. The high agreement of SSHs from this automated approach with those from manual selection indicates the reliability and usefulness of this approach. Within a 45-km section of one ATM L1B file, SSH demonstrates a linear gradient, which is applied to derive SSHs where there are no leads. The resulted SSHs are then used to compute total freeboard and ice thickness. This automated approach is also tried to retrieve SSH and then compute the total freeboard on one entire IceBridge sea ice flight each in Arctic and Antarctica.

Decadal Variations of Sea Ice Thickness in the Amundsen‐Bellingshausen and Weddell Seas Retrieved From ICESat and IceBridge Laser Altimetry, 2003–2017

“…The OIB campaigns have surveyed the Antarctica in October and November since 2009 and partially filled the gap between ICESat, in orbit from 2003 to 2009, and ICESat-2 launched in September 2018. Each year, there were normally three sea ice flights in the Southern Ocean, one in the ABS and two in the Weddell Sea (Kwok & Maksym, 2014;Wang et al, 2013Wang et al, , 2016. The basic information on these OIB sea ice campaigns are summarized in Table 1.…”

“…Data from two OIB airborne instruments are analyzed in this paper, ATM L1B footprint (1 m) elevation data and high-resolution (0.1 m) DMS images (Dominguez, 2017;Studinger, 2013). DMS images are acquired concurrently with ATM and applied to identify the leads, open water or thin sea ice, which are then used as a mask to select the ATM elevation data as sea surface tie points to compute a local instant SSH reference (Wang et al, 2016). ATM L1B data are used to determine SSH and to calculate the total freeboard.…”

“…The second type of approach, called "SILDAMS" (Onana et al, 2013), utilizes DMS images to automatically identify leads and then determine the ATM tie points for calculating SSH. The third approach first uses a dynamic threshold to identify leads from DMS images and then apply the Apparent Reflectivity and Q-test (DARQ) to choose the ATM tie points for calculating SSH (Wang et al, 2016). The modified Q-test is used to remove the outlier points beyond the height range of mean ± one standard deviation of the tie points (Wang et al, 2016).…”

“…The third approach first uses a dynamic threshold to identify leads from DMS images and then apply the Apparent Reflectivity and Q-test (DARQ) to choose the ATM tie points for calculating SSH (Wang et al, 2016). The modified Q-test is used to remove the outlier points beyond the height range of mean ± one standard deviation of the tie points (Wang et al, 2016). We used the DARQ method to process all DMS images and ATM data in this study and to calculate the SSH and total freeboards in the ABS and Weddell Sea from 2009 to 2016/2017.…”

“…These OIB campaigns provide unique opportunities to study the Antarctic sea ice total freeboard and thickness variations along ground tracks at very high precision. Various approaches have been developed to retrieve the total freeboard and sea ice thickness (Kurtz et al, 2013;Kwok et al, 2012;Kwok & Maksym, 2014;Onana et al, 2013;Wang et al, 2013Wang et al, , 2016Yi et al, 2015). What is missing is a comprehensive study to investigate the variation of total freeboard and ice thickness from all OIB campaigns from 2009 to present in the ABS and Weddell Sea, the majority of the OIB sea ice data in Antarctica.…”