Spatio-temporal noise and drift in fiber optic distributed temperature sensing

Abstract

Distributed temperature sensing (DTS) allows for simultaneous measurement at many remote locations along an optical fiber probe and is a valuable tool in a broad range of applications, such as downhole oil production, dike structural monitoring or fire protection. The specific requirements on spatial, temporal and temperature resolution and on absolute measurement uncertainty vary with the applications. We investigate the spatio-temporal noise and drift properties of two exemplary Raman backscatter DTS systems and discuss the effect of spatial and temporal data averaging. An Allan deviation analysis provides insight into the optimal degree of averaging for a given distance range along the fiber probe. A temperature calibration procedure is employed to retrieve the temperature sensitivity of the DTS system and to compensate for the systematic spatial slope of recorded DTS temperature measurement traces. In response to small temperature steps of a thermally homogeneous and stable water bath environment, we observe a temperature resolution of approximately 0.05 °C at a chosen 1000 m sampling distance along the fiber probe.

Double-Ended Calibration of Fiber-Optic Raman Spectra Distributed Temperature Sensing Data

“…Figure 4 shows the results. The average RMSE of the points before the splice, from 326 m to 338 m, was 0.080 K. The RMSE of the points after the splice, from 1,224 m to 1,236 m, was 0.10 K. The overall RMSE of the bath temperature, based on the average of all 15 points, was 0.060 K and the bias was 0.003 K. Under laboratory circumstances, Voigt et al [ 28 ] found a standard deviation that was very close to the factory supplied specification of 0.07 K for measurement integration times of 60 s at 1 km. Here, we find slightly higher values than specified by the factory but this may be partially due to the fact that the accuracy of the TSic 506 measurement was around 0.04 K. We may conclude that also under these field conditions, one can obtain temperature measurements with accuracy close to the nominal one after careful calibration.…”

Stream heat budget modeling with HFLUX: Model development, evaluation, and applications across contrasting sites and seasons

“…This example 502 suggests that while good results can be obtained with remote weather stations, in-situ 503 measurements may provide better results. Another likely explanation for the error in the 504 modeled stream temperatures at Ninemile Creek is the systematic, high frequency noise that is 505 inherent to fiber optic DTS data collected over short integration times and distances (DC3; 506 Briggs et al, 2012;Voigt et al, 2011). Discrepancies exist between the measured temperatures, 507 which include this high frequency noise, and the modeled temperatures.…”

Limitations of fibre optic distributed temperature sensing for quantifying surface water groundwater interactions

“…The results of this study demonstrate that temperature measurements with an accuracy of approximately 0.1 • C can be obtained by fiber optic DTS. A more accurate temperature precision can be obtained with integration, either over longer time or greater lengths of cable (Voigt et al, 2011). However, this would be at the expense of measuring finer spatial scales or more rapid temporal changes.…”