Surveying Infrasonic Noise on Oceanic Islands

An essential step in the establishment of an International Monitoring System (IMS) infrasound station is the site survey. The survey seeks a location with relatively low infrasonic noise and the necessary logistical support. This paper reports results from our surveys of two of the oceanic sites in the IMS - the Azores and Cape Verde. Each survey sampled infrasonic noise, wind velocity, air temperature and humidity for ~3 weeks at 4 sites near the nominal IMS locations. The surveys were conducted on Sao Miguel (the main island in the Azores) and Maio (Cape Verde). Infrasonic noise was measured using the French MB2000 microbarometer.During our 3-week experiment in January the trade winds at Cape Verde varied little from an azimuth of 63°. Because of the unvarying wind azimuth, the experiment gave us an opportunity to examine the effectiveness of a forest at reducing both wind speed and infrasonic noise. We find that the thick Acacia forest on Maio reduces wind speeds at a 2 m elevation by more than 50% but does not reduce infrasonic noise at frequencies below 0.25 Hz. This forest serves as a high-frequency filter and clearly does not reduce long-period noise levels which are due to large-scale turbulence in the atmospheric boundary layer above the forest. This is consistent with our observations in the Azores where the relationship between infrasonic noise and wind speed is more complex due to frequent changes in wind azimuth.In Cape Verde, wind speed and infrasonic noise are relatively constant. The diurnal variations are clearly seen however the microbarom is only rarely sensed. In the Azores, during our 3-week experiment in November and December of 1998, wind speed and infrasonic noise change rapidly. At this location, daily noise level swings of 40 to 50 dB at 0.1 Hz are not uncommon in the early winter and are due to changes in wind speed and atmospheric turbulence. The effectiveness of an infrasound station in the Azores will be strongly dependent on time during the winter season.The two surveys illustrate some of the difficulties inherent in the selection of sites for 1 to 3 km aperture arrays on oceanic islands. Due to elevated noise levels at these sites, 8 element, 2 km aperture arrays are strongly preferred.