Investigate daily average (noise) amplitudes at a given frequency for every data channel. The daily average is corrected (levelled) by the median of average values of all stations. These levelled daily average values stay normally within a certain range. Unusual deviations of this value detected in a long term statistics indicate problems with data or metadata. The following table shows automatically detected deviations for three channels (ZNE or Z12), if existing, in color codes: green - OK, yellow - the last day shows unusual deviation, orange - the last week shows unusual deviation, red - the last month shows unusual deviation, gray or white - no (recent) data. The station set is separated in broadband stations (VBB) and short period (SP) stations, investigated at different frequencies. VBB stations show low scatter in the noise data at 0.15 Hz since the signal is dominated by the ocean microseism. For SP stations this frequency is out of scale. Therefore a higher frequency is chosen where the scatter is much larger and the resolution for detecting errors is reduced.
The figures below the table show the individual measurements on the channels.
Description see there.
This page is daily updated.last 365 days, created 21-Sep-2021
VBB Stations @ 0.15Hz
SP Stations @ 1.00Hz
Colors: no data amplitude ok last day over thresh last week over thresh last month over thresh
The figures below show the levelled average of the last 365 days (x-axis: 0 = one year ago, 364 = yesterday) in color codes, scaled by the individual standard deviation of the measurement series. Bold colors in red and blue indicate large deviations from the average to high or low values, respectively. Gray or pastel shaded colors represent the normal scatter. White means no data computed. The numbers after 'm:' and 's:' in the figure titles are the relative mean and the standard deviation in log10(nm/s), respectively. Please note that the mean and standard deviation is computed after removing the upper and lower 15% of the data to remove the influence of outliers. Maximum color boldness is reached at 10 standard deviations (red for positive, blue for negative amplitudes). The color intensity for log amplitudes below 5.5 standard deviations is reduced by a square function. Similar color patterns at neighbored stations indicate natural undulations in the noise signal.