| The Rezatec Dam Monitoring product highlights statistical anomalies. It is for the dam owner to interpret these anomalies in context and decide if further action needs to be taken. Ground Motion anomalies An anomaly is defined as an observation that is outside at the extreme of the distribution of movement for a particular location. Given that the trend provides the average value of movement for each time step, a 95% confidence interval is used to set a threshold. This means the 5% of the outlying observations are identified as anomalies. The below chart shows a section of dam with high variability, the anomalies are marked in red.  Anomalies are then counted over 3 rolling time periods: 30 days, 1 year, 3 year The values shown on the map and in the summary table are the total number of anomalies in the given time period. Vegetation anomalies In the vegetation layer, anomalies are defined in a very similar way. The only difference is that here it is expected that time-series present a strong seasonal component, which means that big swings can be common. This makes the use of the previous method unviable. A different algorithm has therefore been used, based on time-series decomposition, which simply means that it defines anomalies after the seasonal and trend components have been removed. Below is an image of an output of this algorithm applied to the dataset used in the value showcase:  |