Avalanche forecasting — an expert system approach
Abstract Avalanche forecasting for a given region is still a difficult task involving great responsibility. Any tools assisting the expert in the decision-making process are welcome. However, an efficient and successful tool should meet the needs of the forecaster. With this in mind, two models, wer...
| Published in: | Journal of Glaciology |
|---|---|
| Main Authors: | , |
| Format: | Article in Journal/Newspaper |
| Language: | English |
| Published: |
Cambridge University Press (CUP)
1996
|
| Subjects: | |
| Online Access: | http://dx.doi.org/10.1017/s0022143000004172 https://www.cambridge.org/core/services/aop-cambridge-core/content/view/S0022143000004172 |
| Summary: | Abstract Avalanche forecasting for a given region is still a difficult task involving great responsibility. Any tools assisting the expert in the decision-making process are welcome. However, an efficient and successful tool should meet the needs of the forecaster. With this in mind, two models, were developed using a commercially available software: CYBERTEK-COGENSYS TM , a judgment processor for inductive decision-making–a principally data-based expert system. Using weather, snow and snow-cover data as input parameters, the models evaluate for a region the degree of avalanche hazard, the aspect and altitude of the most dangerous slopes. The output result is based on the snow-cover stability. The new models were developed and have been tested in the Davos region (Swiss Alps) for several years. To rate the models, their output is compared to the a posteriori verified hazard. the first model is purely data-based. Compared to other statistical models, the differences are: more input information about the snow cover from snow profiles and Rutschblock tests, the specific method to search for similar situations, the concise output result and the knowledge base that includes the verified degree of avalanche hazard. The performance is about 60%. The second, more-refined model, is both data- and rule-based. It tries to model the decision-making process of a pragmatic expert and has a performance of about 70%, which is comparable to the accuracy of the public warning. |
|---|