100 Ocean Bottom Nodes With Their Csac Clock Drift Analysed For 91 Days Three Controlled Experiments

Summary

OBN survey numbers have been steadily increasing over recent years. However, those surveys come with new challenges compared to streamer or cable acquisitions. One of the main challenges is that most nodes are autonomous and cannot communicate with external devices to obtain frequent, exact GPS timing during the production period. Instead, nodes use an internal clock to time the recorded data, which over the course of the production period will deviate from GPS time for any node. This is known as clock drift and is usually compensated for in the processing of the data. How well different correction methods work depends on the stability and predictability of the drift. We have conducted a series of experiments aiming to characterize the main drift elements of CSAC clocks used in some commercial OBN, namely disciplining, aging and temperature. We run controlled drift tests with and without disciplining for 60, 75 and 91 days respectively, using 100 nodes. We show that nodes without clock disciplining can be equally well drift corrected as their disciplined counterparts. We further show that residuals from linear correction can be fully attributed to aging as a dominant effect – there is no chaotic behaviour in our recorded data. In fact, we observe that the clocks are very stable and to a high degree predictable. In a separate experiment, we also investigated the effects of a temperature changes. The results show that this is causing mostly a small linear drift element which the linear corrections are taking care of.