CLIMIS4AVAL

Introduction:

The backbone of the Swiss avalanche forecast infrastructure are the data of the Intercantonal Measurement and Information System (IMIS), which currently consists of 187 automated snow and weather stations. They are distributed throughout the Swiss Alps and in most cases are situated in the high alpine region above the tree line. The stations continuously record the snow and weather conditions, transmit data to the SLF every hour and provide the national avalanche warning service of the SLF as well as local avalanche services responsible for public safety in settlements and on roads with the crucial information for danger assessment. Public products such as new snow or snow height maps also rely on these data.

Therefore, the objective is the development of algorithms that allow real-time detection of anomalies in the time series, but also the detection of outliers, and impute missing data by applying state-of-the-art machine learning approaches.

This real-time data cleansing will solve a long-standing issue with the IMIS data that are known to be contaminated with data anomalies and has hindered automated processing. Hence, the completion of the proposed research will have a major impact, in particular for the application of numerical avalanche prediction models such as we recently developed in collaboration with the SDSC.

Problem:

This project aims to perform anomaly detection, outlier detection, and imputation on IMIS station data both retrospectively and in real time. In particular, the project focuses on the measurements of snow depth, wind speed and direction, air temperature, precipitation, and a maintenance model.

Proposed approach:

The SDSC is working with SLF to apply time-series-based statistical and machine learning-based methods for anomaly detection, outlier detection, and imputation in the given context.

Impact:

Numerical avalanche prediction and other models used at SLF will be more accurate with the cleaned data. The cleaned data will also be made openly available in the data portal of SLF, and will therefore benefit the numerous downstream users of the data.

‍