Assessment of a vertical high-resolution distributed-temperature-sensing system in a shallow thermohaline environment

Abstract

In shallow thermohaline-driven lakes it is important to measure temperature on fine spatial and temporal scales to detect stratification or different hydrodynamic regimes. Raman spectra distributed temperature sensing (DTS) is an approach available to provide high spatial and temporal temperature resolution. A vertical high-resolution DTS system was constructed to overcome the problems of typical methods used in the past, i.e., without disturbing the water column, and with resistance to corrosive environments. This system monitors the temperature profile each 1.1 cm vertically and in time averages as small as 10 s. Temperature resolution as low as 0.035 °C is obtained when the data are collected at 5-min intervals. The vertical high-resolution DTS system is used to monitor the thermal behavior of a salt-gradient solar pond, which is an engineered shallow thermohaline system that allows collection and storage of solar energy for a long period of time. This paper describes a method to quantitatively assess accuracy, precision and other limitations of DTS systems to fully utilize the capacity of this technology. It also presents, for the first time, a method to manually calibrate temperatures along the optical fiber

Calibrating Single-Ended Fiber-Optic Raman Spectra Distributed Temperature Sensing Data

“…At position z (m), the power of the Raman Stokes, P S ( z ), and anti-Stokes, P aS ( z ), signals are translated into temperatures according to [16]: $T\left(z\right)=\frac{\gamma }{\text{ln}\frac{P_S\left(z\right)}{P_{aS}\left(z\right)}+C-\mathrm{\Delta }\alpha z}$where γ (K) represents the shift in energy between a photon at the wavelength of the incident laser and the scattered Raman photon, C is a dimensionless calibration parameter that encompasses properties of the incident laser and the DTS instrument itself, and Δα (m −1 ) is the differential attenuation between the anti-Stokes and Stokes signals in the fiber. A complete derivation of this equation can be found in the online supplemental material.…”

“…When two reference points with different temperatures are available, these parameters can be calculated explicitly by simultaneously solving Equation (1) at each reference point [16]. If two different reference temperatures are not available (or the two reference points are at similar temperatures), the parameters must be determined through optimization.…”

“…For each reference section, representative values for distance ( z* ) and log power ratio ( $\text{ln}\left[\frac{P_S*}{P_{\mathrm{italicAS}}}\right]$) must be determined. As the inverse of temperature varies linearly with both z and $\text{ln}\left[\frac{P_S}{P_{\mathrm{italicAS}}}\right]$ (Equation (1)), the representative values z* and $\text{ln}\left[\frac{P_S*}{P_{\mathrm{italicAS}}}\right]$ are best calculated as the arithmetic mean of each of these parameters along the reference section: $z*=\frac{1}{n}\sum _{i=1}^n{z}_i$ $\text{ln}\frac{P_S*}{P_{aS}}=\frac{1}{n}\sum _{i=1}^n\text{ln}\frac{P_S(z_i)}{P_{aS}(z_i)}$Using reference sections rather than single points can greatly improve both the precision and accuracy of the recalibrated data [16]. Tyler et al [3] recommend reference sections of at least ten times the sampling interval for calibration purposes.…”

“…To have controlled conditions, the solar pond was built inside a laboratory and was subject to high-density discharge lamps that were turned on for 12 hours per day [20]. The pond was instrumented with a variety of sensors to monitor its performance, including a vertical high-resolution DTS system [16]. This laboratory setup provided nearly ideal conditions for a DTS installation, with four calibration baths in a well-controlled environment.…”

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

“…The advent of new DTS applications was made possible by the development of robust fiber optic cables and improved specifications and lower costs of DTS systems. With good calibration, even less expensive DTS systems provide 1–2 m resolutions along cables of up to 5 km, with 0.1 K accuracies for integration times of 60 s. Applications started with dam surveillance [3] and borehole observations [4] and have multiplied rapidly to include monitoring of ice caves [5], estuaries [6], sewers [7], solar ponds [8], soils [9–11], groundwater [12,13], streams [14–17], lakes [18], atmosphere [19,20], electric transmission cables [21], mines [22], and gas pipelines [23]. …”