DFG Exploring Aufeis
Relevance of Icing and Ice Reservoirs for Climate Change Adaptation in the Trans-Himalaya of Ladakh, India
Principal Investigator:
Prof. Dr. Marcus Nüsser
Project Team:
Dr. Dagmar Brombierstäudl
Dr. Susanne Schmidt
Tobias Schmitt, B.Sc.
Dr. Mohd Soheb
Period: 2021-2024
Foundation: Deutsche Forschungsgemeinschaft (DFG)
Project Number: NU102/15-1
"Aufeis is the most neglected cryospheric component in the cold-arid regions of High Asia."
What is aufeis?
Aufeis (also known as icing or naled in Russian) is a common phenomenon in many cold and permafrost regions. It describes seasonal ice masses that form during winter and result from successive freezing of water that seeps from the ground, originates from springs, or emerges through cracks in river and lake ice. This icing process results in a layered ice body that stores the winter discharge, which is subsequently released in spring, contributing to river flow. Aufeis fields vary considerably in size and can significantly contribute to annual run-off in affected environment. They often reappear in the same locations each year. Their formation and distribution depend on various factors, such as topography, climate, permafrost, and snow cover, but these interactions are still poorly understood. In Ladakh, aufeis occurrence has been document in historical reports and can be witnessed on satellite imagery and field observations in many tributaries.
Different types of aufeis
Different types of ice reservoirs
Due to the cold-arid climatic conditions, agriculture depends on irrigation systems (Nüsser et al. 2012). To bridge critical periods of water scarcity in spring caused by the late onset of glacier melt (Schmidt & Nüsser 2017, 2023) and snowless winters (Passang et al. 2023), ice reservoirs were introduced in various tributaries since the 1980s, commonly referred to as “artificial glaciers” (Nüsser & Baghel 2016, Nüsser et al. 2019a, b). These structures capture winter runoff as ice close to the cultivated areas utilizing the natural aufeis (icing) process, which occurs in many of the upper tributaries in the region. Since the 1980s, different types of ice reservoirs have been developed. The first ice reservoirs were constructed along the streams, where a cascade of walls reduces the flow velocity. Later, the diversion type was introduced to divert water to shallow grounds. In the most recent structures, ice is stored vertically.
How did we study aufeis in the Trans-Himalaya?
Mapping the spatial distribution
To understand aufeis distribution in the Trans-Himalaya, we analyzed freely available optical satellite imagery from the Landsat and Sentinel-2 missions from 2008 onwards. To complement the resulting inventories, we derived topographical parameters from digital elevation models. For our studies, we utilized a range of remote sensing methods. In the first study, we characterised the seasonality of aufeis development in the Upper Indus Basin with a harmonic time series and mapped aufeis fields with a semi-automatic threshold approach (Brombierstäudl et al. 2021). In a second study, we conducted a detailed spatio-temporal assessment of icing and aufeis patterns in the Tso Moriri Basin. Here we used a machine learning classifier (Random Forest) to automatically map water overflow and resulting aufeis fields (Brombierstäudl et al. 2023). This sequential application of time series and machine learning provided us with crucial information about a region-specific aufeis seasonality. This allowed us to map aufeis in the Pangong Tso basin with a small amount of snow-free satellite imagery in the third study (Schmitt et al. 2024). All results were validated with visual interpretation of high-resolution satellite images and during field surveys in winter.
Measuring volumes and thickness
To estimate aufeis volumes and thickness, we used very high-resolution Pléiades satellite imagery and terrestrial photographs taken during fieldwork at selected sites in summer and winter. In total, we surveyed ice reservoirs in four tributaries (Phuktse, Igoo, Mathoo, and Nang) and natural aufeis sites in two tributaries (Stok and Gya). Satellite imagery was processed for Phuktse, Igoo, and Gya. We created 3D models from the overlapping 2D imagery with the computer vision and photogrammetric technique Structure-from-Motion. Terrestrial photographs were processed with the software package Agisoft Metashape Professional, and satellite imagery with the NASA Ames Stereo Pipeline. All resulting point clouds were rasterized at a spatial resolution of 2 m to obtain digital elevation models. A raster-based approach of digital elevation model differencing of the summer from the winter DEM was applied to quantify ice thickness (Brombierstäudl et al. 2024). In addition, we measured the aufeis density and depth at each site to estimate the ice water equivalent.
Measuring aufeis thickness and water equivalent in winter
terrestrial photography
Key findings
Aufeis distribution and temporal patterns in the Trans-Himalaya
We detected a total of ~ 400 km² of aufeis across three basins: Upper Indus Basin: 298 km² (Brombierstäudl et al. 2021), Tso Moriri Basin: 9 km² (Brombierstäudl et al. 2023), and Pangong Tso Basin: 86 km² (Schmitt et al. 2024). The area was spread across 4887 individual aufeis fields. While smaller aufeis fields (< 0.1 km²) were more common, larger aufeis fields (> 0.1 km²) accounted for 75 % of the total aufeis-covered area. These large aufeis fields are typically found along braided river systems. One of the largest aufeis fields measured 14 km² and is located in the Pangong Tso Basin (Schmitt et al. 2024), which is almost three times larger than the largest high-altitude glaciers in Ladakh (Schmidt & Nüsser, 2017). The regional distribution revealed a distinct longitudinal gradient with increasing number and size of individual aufeis fields towards the Tibetan Plateau. This underscores the importance of continental and cold-arid climatic conditions for the icing process. Aufeis in the region typically is formed at elevations between 4000 to 5000 m a.s.l., suggesting that thermal and hygric factors limit its formation beyond these elevations. Aufeis fields of various sizes are distributed across all elevation bands, indicating no direct relation between elevation and size. It primarily occurs along streams with gentle gradients < 10° where reduced flow velocity and turbidity of water promote freezing (Brombierstäudl et al. 2021, 2023; Schmitt et al. 2024).
Temporal patterns
Aufeis accumulation starts in November and reaches its peak between January and February. Melting commences in April and continues until July, with most aufeis disappearing by August. Freezing is driven by regular water overflow from either surface water channels, seepages through the riverbed alluvium, or from wetland areas. Similar to other aufeis-affected environments, aufeis formation occurs at the same locations each year. Both, aufeis accumulation and melting exhibit seasonal and inter-annual variability that mainly depends on elevation, climatic conditions, and water availability in a given year (Brombierstäudl et al 2023).
Seasonal evolution of aufeis in the Tso Moriri Basin
Aufeis volumes and thickness
Quantification of aufeis volumes and thickness in the season 2022/23 through digital elevation model differencing revealed substantial amounts of ice volumes ranging from 44,969 m³ in Phuktse Valley up to 105,790 m³ in Sasoma Valley, which corresponds to an ice water equivalent of 35,525,510 l (Phuktse) to a maximum of 83,574,100 l (Sasoma). Aufeis thickness at natural aufeis sites exceeds 3 m. Even though the median ice thickness accumulated in the ice reservoirs is lower compared to its natural counterparts, almost similar ice thickness is reached in the upper sections of both ice reservoirs. Here, the small stone walls are partly overflown by ice, which demonstrates their efficacy for aufeis accumulation. In addition, the uneven distribution of ice thickness presumably indicates locations of water discharge onto the surface (Brombierstäudl et al. 2024).
References
BROMBIERSTÄUDL D, SCHMIDT S, SOHEB M & NÜSSER M (2024): Aufeis thickness and volume estimations from stereo satellite imagery and terrestrial photographs: Evidence from Central Ladakh, India. Science of The Total Environment 954, 176180. doi: https://10.1016/j.scitotenv.2024.176180
BROMBIERSTÄUDL D, SCHMIDT S & NÜSSER M (2023): Spatial and temporal dynamics of aufeis in the Tso Moriri basin, eastern Ladakh, India. Permafrost and Periglacial Processes 34(1): 81-93. doi:10.1002/ppp.2173
BROMBIERSTÄUDL D, SCHMIDT S & NÜSSER M (2021): Distribution and relevance of aufeis (icing) in the Upper Indus Basin. In: Science of The Total Environment 780. doi: 10.1016/j.scitotenv.2021.146604
NÜSSER M & BAGHEL R (2014) The Emergence of the Cryoscape: Contested Narratives of Himalayan Glacier Dynamics and Climate Change. In: Schuler B (ed) Environmental and Climate Change in South and Southeast Asia. Brill, Leiden, pp 138–156
NÜSSER M, DAME J, KRAUS B, BAGHEL R & SCHMIDT S (2019a) Socio-hydrology of “artificial glaciers” in Ladakh, India: assessing adaptive strategies in a changing cryosphere. Reg Environ Change 19:1327–1337. https://doi.org/10.1007/s10113-018-1372-0
NÜSSER M, DAME J, PARVEEN S, KRAUS B, BAGHEL R & SCHMIDT Sl (2019b) Cryosphere-Fed Irrigation Networks in the Northwestern Himalaya: Precarious Livelihoods and Adaptation Strategies Under the Impact of Climate Change. Mt Res Dev 39:R1–R11. https://doi.org/10.1659/MRD-JOURNAL-D-18-00072.1
NÜSSER M, SCHMIDT S & DAME J (2012) Irrigation and Development in the Upper Indus Basin: Characteristics and Recent Changes of a Socio-hydrological System in Central Ladakh, India. Mt Res Dev 32:51–61. https://doi.org/10.1659/MRD-JOURNAL-D-11-00091.1
SCHMIDT S & NÜSSER M (2017) Changes of High Altitude Glaciers in the Trans-Himalaya of Ladakh over the Past Five Decades (1969–2016). Geosciences 7:. https://doi.org/10.3390/geosciences7020027
SCHMITT T, BROMBIERSTÄUDL D, SCHMIDT S & NÜSSER M (2024): Giant Aufeis in the Pangong Tso Basin: Inventory of a Neglected Cryospheric Component in Eastern Ladakh and Western Tibet. Atmosphere 15(3):236. doi:https://doi.org/10.3390/atmos15030263