Correction: O’Brien et al. Correction of Radiometry Data for Temperature Effect on Dark Current, with Application to Radiometers on Profiling Floats. Sensors 2022, 22, 6771

The authors wish to correct the following errors in the original paper [...].


Text Correction
The following correction has been made to the Abstract: The corrections are based on modeling the temperature of the radiometer and show an average bias in the measured value of nearly 1 × 10 −4 W m −2 nm −1 , an order of magnitude larger than the reported uncertainty of 2.5 × 10 −5 W m −2 nm −1 for the sensors deployed on BGC-Argo floats (SeaBird scientific OCR504 radiometers).
In the subsection Profile Extraction, Quality Control and Modeling under Section 2: Materials and Methods, in the sixth paragraph, we make the correction: We constrained measurements to the range |E d | < 3 × 10 −4 W m −2 nm −1 and |PAR| < 0.5 µmol photons m −2 s −1 .
In the subsection Profile Extraction, Quality Control and Modeling under Section 2: Materials and Methods, in the seventh paragraph, we make the correction: This test is important as dEd/dTs is small (−3 × 10 −5 to 3 × 10 −5 W m −2 nm −1 • C −1 ) and hence not detectable relative to other environmental processes if the temperature gradient in a profile is too small.
In Section 3: Results, in the second paragraph, we make the correction: (Figure 1).
In Section 4: Discussion and Summary, in the first paragraph, we make the following correction: For this BGC-Argo dataset, the mean absolute temperature corrections on E d using night and day profiles are 8 × 10 −5 and 9.3 × 10 −5 (W m −2 nm −1 ) and maximum absolute corrections are 4.4 × 10 −4 and 6.14 × 10 −4 (W m −2 nm −1 ), respectively (Table 2). These corrections are more than an order of magnitude larger than the known sensitivity of the sensors (2.5 × 10 −5 W m −2 nm −1 ).
In Section 4: Discussion and Summary, in the fourth paragraph, we make the following correction: They provide a model that includes drift correction, where in certain, but rare, cases they observed a drift as much as 1 × 10 −7 days −1 , producing a significant correction over a three-year lifetime.
In Section 4: Discussion and Summary, in the fourth paragraph, we make the following correction: The values of our modeled coefficients dE dark /dT agree well with [4] and [5], with the maximum dE dark /dT on the order of 2 − 4 × 10 −5 W m −2 nm −1°C−1 (Figure S9 in [5]).
In Section 4: Discussion and Summary, in the fourth paragraph, we make the following correction: For dPAR dark /dT, ref.
[4] produces the smallest values, on the order of 2 × 10 −2 µmol photons m −2 s −1°C−1 , while [5] agrees with our maximums as high as 4 × 10 −2 µmol photons m −2 s −1°C−1 . Likewise, we find a similar model constant of PAR (our x 0 ) with [5] showing the highest maximum. Our investigation of the daytime profiles revealed these significant dark readings at depth, and our corrections for PAR are of the same order relative to surface values as our corrections for E d : at 10 m, our average PAR correction is on the order of 0.001% of the 10 m measured PAR value, analogous to the average 10 m correction at all three wavelengths. Overall we find very similar results between our method and [4,5].

Error in Table
We update Tables 1-3 from the original publication to adjust all values by the scaling factor of 100 by adjusting the presented values to the reported units. The corrected tables appear below: Table 1. Nonzero x 1 by the night method and day method for all λ (W m −2 nm −1°C−1 or µmol photons m −2 s −1°C−1 ), as shown in Figure 2.

Error in Figure
We update Figures 1-6

Error in Figure
We update Figures 1-6 from the original publication to adjust all values by the scaling factor of 100 by adjusting the presented values to the reported units. The corrected figures appear below.               We apologize for the inconvenience or confusion caused by our error in the original manuscript. The changes do not affect the method presented, but they do affect the magnitude of the bias identified in the sensors. The authors state that the scientific conclusions are unaffected. This correction was approved by the Academic Editor. The original publication has also been updated.