About the project
The snow, permafrost, and glaciers that form the terrestrial cryosphere play a key role in the climate system by regulating the global cycles of energy, water, and carbon. As such, all of these cryospheric components have been designated the status of Essential Climate Variables (ECVs) by the Global Climate Observing System (GCOS). Nonetheless, these ECVs are only partially observable using Earth Observation (EO) satellites due to spatio-temporal gaps, sensor noise, and the indirectness of retrieval algorithms.
As demonstrated through the initialization of numerical weather predictions and the generation of atmospheric reanalyses, data assimilation presents a natural solution to this problem of partial observability. Data assimilation has nonetheless received relatively little attention from the EO community despite its potential for adding value to satellite data by both filling spatio-temporal gaps and inferring latent ECVs with uncertainty quantification.
In the project PATCHES we aim to apply ensemble-based data assimilation techniques to improve on existing methods to infer cryospheric ECVs from EO. The main focus is on seasonal snow where accurate snow mass (i.e. snow water equivalent) estimation in mountains remains a central unsolved problem.
Objectives
The primary objective of PATCHES is to generate a high resolution global ensemble-based seasonal snow mass reanalysis by leveraging the multi-decadal EO snow data from the ESA Snow_cci project. The envisaged ensemble-based reanalysis workflow in PATCHES can be adapted to permafrost and glaciers, as well as other terrestrial ECVs.
A secondary objective is to attempt to exploit repeated snow patterns in historical satellite imagery so as to downscale the coarser resolution but long term Snow_cci products from AVHRR with probabilistic machine learning techniques.
Background
The engine driving the PATCHES project is the mathematical framework known as data assimilation (or inverse modeling) in the geosciences. For a relatively recent overview of this method and its application to snow see:
The Multiple Snow Data Assimilation System (MuSA v1.0)
The project is also relying on synergies with the in-house CryoGrid community model by adopting workflows used for its application to global permafrost reanalyses in the ESA Permafrost_cci project.
Moreover, we will make use of the largely untapped snow climate data record from the AVHRR sensors developed in the Snow_cci which, despite their coarser resolution, provide near daily global fractional snow-covered area retrievals since the dawn of the satellite era.
PATCHES is also leveraging local UiO expertise in global glacier modeling, seasonal snow, climate data, remote sensing, and emerging observing systems. Beyond UiO, in PATCHES we are working together with partners with expertise in snow remote sensing (Uni. Bern), snow modeling (IPE), and topography-based downscaling (SLF).
Financing
The PATCHES project is funded by the European Space Agency (ESA) Climate Change Initiative (CCI) program as an ESA CCI Research Fellowship, granted to Kristoffer Aalstad, researcher at Department of Geosciences, University of Oslo.
The project is for 2-years, it started up in October 2023 and wll end in October 2025.
Cooperation
The PATCHES project is carried out at the Section for Geography and Hydrology (GeoHyd) at University of Oslo, in close cooperation with Permafrost_cci, the Snow_cci, and with researchers from the University of Oslo hosted research projects; the GLACMASS project (ERC), the SNOWDEPTH project (RCN), and the ACTIVATE project (ERC).
The PATCHES project is also connected to the cross-disciplinary Land-ATmosphere Interactions in Cold Environments (LATICE) research initiative at UiO, the snow research group at GeoHyd, and external partners at University of Bern, IPE, and SLF.