Abstract
We present an approach for extracting quantifiable information from archival aerial photographs to extend the temporal record of change over a region of the central eastern Greenland Ice Sheet. The photographs used were gathered in the 1930s as part of a surveying expedition and were not originally acquired with photogrammetric analysis in mind. Nevertheless, we make opportunistic use of this imagery, alongside additional novel datasets, to explore changes at ice margins well before the advent of conventional satellite technology. This longer record is crucial for improving our understanding of how glaciers respond to a changing climate. Our work focuses on a series of relatively small outlet glaciers from the eastern margin of the Ice Sheet. We demonstrate that while air and sea surface temperatures are important controls, there is significant heterogeneity in glacial responses, with non-climatic controls—such as fjord bathymetry—playing an extremely important role. We suggest that focusing solely on regional variations masks important variability; a full understanding of the Ice Sheet requires considering changes at the scale of individual outlet glaciers.
Introduction: Overcoming the Satellite-Era Bias
While it is well-established that the Greenland Ice Sheet (GrIS) is losing mass at an accelerating rate, our high-resolution observations are largely limited to the "satellite era," which began in the 1970s. This temporal limitation creates a bias, making it difficult to distinguish long-term climate-driven trends from short-term natural fluctuations. Furthermore, glaciological research often prioritizes massive outlet glaciers like Jakobshavn or treats entire regions as homogeneous blocks. This study addresses the "local heterogeneity" problem: the reality that smaller outlet glaciers, which constitute a significant portion of the ice sheet's margin, often exhibit unique, localized responses to warming that regional averages tend to obscure.
Methodology: Structure-from-Motion and Archival Data
To extend the record back by 40 years, researchers employed Structure-from-Motion (SfM) photogrammetry to breathe new life into archival data. The primary source was the British Arctic Air Route Expedition (BAARE) of 1930–31, which captured oblique aerial photographs for basic surveying. By processing these into georeferenced, map-accurate orthophotos and aligning them with modern digital elevation models (ArcticDEM), the team tracked glacier terminus positions across seven discrete time steps spanning 89 years.
This archival data was synthesized with several other datasets to create a century-long timeline:
- CORONA (1966): Declassified US spy satellite imagery.
- Greenland Survey (1978–87): Vertical aerial photography from the Geodetic Institute of Denmark.
- Landsat (1985–2019): Standard multi-spectral satellite imagery.
- Environmental Proxies: 100+ years of air temperature data from the Tasilaaq station, sea surface temperatures (SST), and Surface Mass Balance (SMB) models.
Key Insights: The Role of Bathymetry and Local Factors
The study reveals that regional trends often hide contradictory local behaviors. For instance, while Southeast Greenland is generally thinning, one glacier might retreat rapidly while its neighbor remains stable. This is often due to Bathymetric Gatekeeping. Glaciers situated on shallow seafloor ridges are "pinned" and remain stable despite rising temperatures. However, once they retreat off these ridges into deeper water, they reach a tipping point and can collapse rapidly.
Additionally, the data highlighted a "Mid-Century Pause"—a period of relative stability or slight advance between the 1960s and 1990s. This coincided with a minor dip in regional temperatures and positive surface mass balance before the current era of accelerated retreat began. The study also confirmed that marine-terminating glaciers are far more dynamic and sensitive than land-terminating ones, as they are vulnerable to warming ocean currents from below.
Results: A Century of Retreat
Every glacier studied showed an overall retreat between 1930 and 2019. In the southern study region, some glaciers retreated by as much as 3 km. The most dramatic changes occurred recently; between 2015 and 2019, retreat rates accelerated significantly. A standout example is Glacier 22 (KVJ Steenstrup Nordre Brae), which exhibited a staggering retreat rate of nearly 1 km per year during the final study period.
In the northern region, the data showed even more complexity. While the long-term trend was retreat, a significant number of glaciers actually showed substantial advance between 1985 and 1995. This highlights how local topographic and bathymetric factors can temporarily override global warming trends.
Conclusion and Future Work
The primary challenge remains the lack of comprehensive bathymetric data for remote fjords. While the influence of the seafloor can be inferred, actual depth measurements are needed to confirm these "tipping points." Future research should also shift focus toward surface elevation change (thinning) rather than just frontal position. Because glaciers often thin significantly before their terminus finally collapses, monitoring "dynamic thinning" could provide an earlier warning system for glacial instability across the Greenland Ice Sheet.
Summary updated