The UTC date of each 15-minute park phase observation; yyyy/mm/dd

Pressure acquired with Sea-Bird Scientific SBE 41CP CTD. This is reported as the instrument output with the factory calibration applied.

Salinity acquired with Sea-Bird Scientific SBE 41CP CTD. This is reported as the instrument output with the factory calibration applied.

Temperature at dissolved oxygen sensor; Acquired with Sea-Bird Scientific SBE 63 Optical Dissolved Oxygen Sensor. Temperature at the dissolved oxygen sensor was calculated from the raw SBE 63 thermistor voltage by applying the factory calibration:

L = ln (100000 * thermistor voltage / (3.3 - thermistor voltage)) Temperature [ deg C] = 1 / (TA0 + TA1*L + TA2*L2 + TA3*L3) - 273.15

See https://www.bco-dmo.org/dataset/728371 for calibration coefficients.

Dissolved oxygen; Acquired with Sea-Bird Scientific SBE 63 Optical Dissolved Oxygen Sensor. Dissolved oxygen concentration [umol/kg] was calculated and corrected for salinity and pressure effects according to the 2016 Argo recommendations for processing of dissolved oxygen data (Thierry et al. 2016). As per these recommendations, an initial pressure correction was made to the raw measured oxygen phase delay Phi [usec] according to Bittig et al. 2015:

                  Phi adj = Phi raw + Pcoef1 * P / 1000

where Pcoef1 = 0.115 usec and P [dbar] is the collocated SBE 41CP CTD pressure measurement. The adjusted oxygen phase delay [usec] was then converted to voltage by applying the factory conversion:

                  V = Phiadj / 39.457071

Uncorrected dissolved oxygen concentration [mL/L] was calculated by applying the factory calibration:

                  Oxygen_uncorr [mL/L] = {(A0 + A1*T + A2*V2) / (B0 + B1*V) - 1.0} / (C0 + C1*T + C2*T2)

where T is SBE 63 temperature [deg C] and V is adjusted SBE 63 phase delay [V]. The oxygen calibration coefficients (A0, A1, A2, B0, B1, C0, C1, C2) are provided in Table 1. Dissolved oxygen concentration [mL/L] was then corrected for salinity and pressure effects according to the 2016 Argo recommendations.

                  Oxygen_corr [mL/L] = Oxygenun corr [mL/L] * Scorr * Pcorr

where

                  Scorr = A(T,S,Spreset) * exp(S*(SolB0 + SolB1*Ts + SolB2*Ts2 + SolB3*Ts3) + SolC0*S2)

                  A(T,S,Spreset) = (1013.25 - pH2O(T,Spreset)) / (1013.25 - pH2O(T,S))

                  pH2O = 1013.25 * exp(D0 + D1*(100 / (T + 273.15)) + D2*ln((T + 273.15) / 100) + D3*S)

                  Spreset = 0

                  TS = ln [(298.15 - T) / (273.15 + T)]

T and S are the collocated temperature and salinity measurements from the SBE 41CP CTD, respectively. Salinity correction coefficients are from Benson and Krause 1984 (SolB0 = -6.24523e-3, SolB1 = -7.37614e-3, SolB2 = -1.03410e-3, SolB3 = -8.17083e-3, SolC0 = -4.88682e-7) and pH2O coefficients are from Weiss and Price 1980 (D0 = 24.4543, D1 = -67.4509, D2 = -4.8489, D3 = -5.44e-4). The pressure correction factor Pcorr is calculated as outlined in Bittig et al. 2015 as

                  Pcorr = 1 + (Pcoef2 * T + Pcoef3) * P / 1000

with Pcoef2 = 0.00022 and Pcoef3 = 0.0419. Dissolved oxygen concentration [mL/L] was converted to dissolved oxygen concentration [umol/kg]:

                  Oxygen [umol/kg] = Oxygen [mL/L] * (44.6596 umol/mL) / (ρTheta/1000)

where ρθ is the potential density of seawater [kg/m3] at zero pressure and the potential temperature calculated from collocated SBE 41CP CTD salinity, temperature, and pressure measurements using the pden function in the SEAWATER Matlab library (Morgan and Pender 1993). The value of 44.6596 umol/mL is derived from the molar volume of oxygen gas at standard temperature and pressure, 22.3916 L/mole (e.g., García and Gordon 1992).

See https://www.bco-dmo.org/dataset/728371 for calibration coefficients.

Chlorophyll-a; Acquired with WET Labs MCOMS Chlorophyll Fluorometer (excitation 470 nm/emission 695 nm). A dark offset was subtracted from the raw sensor counts and the result was multiplied by a factory-determined scale factor to obtain fluorometric chlorophyll-a concentration [ug/L]. The dark offset was computed separately for each sensor as the mean of the deep-water minima measured during all profiles between July and September 2013 at BATS (https://www.bco-dmo.org/dataset/728371). Chl [ug/L] = Scale Factor * (Output - Dark Counts)

Calibrated chlorophyll-a; Corrected fluorometric chlorophyll-a concentrations (chl_corr derived above) were calibrated to chlorophyll-a concentrations measured by HPLC of bottle samples collected during concurrent Bermuda Atlantic Time-series Study cruises (available at http://www.bco-dmo.org/project/2124 or http://bats.bios.edu).

Linear regression yields the relationship:

chl_cal [µ=ug/L] = chl [ug/L] * 0.4756 + 0.0064; R2 = 0.92.

See https://www.bco-dmo.org/dataset/728371 for bottle calibration data.

Corrected particulate backscattering coefficient bbp (700nm); Depth profiles of particulate backscattering coefficient bbp(700 nm) were despiked using a running median filter (Briggs et al. 2011). Due to the size of the median filter window the initial and final six data points in each profile could not be despiked and appear as NaN. Park phase bbp(700 nm) data were not despiked because they were collected at a nominally constant depth. Despiked profile phase bbp(700 nm) and raw park phase bbp(700 nm) were further corrected by a float-specific deep-water offset. The minimum bbp(700 nm) measured during the long-term deployments was determined for each float (-2.0488e-04 m-1 for F033; -1.1139e-04 m-1 for F034) and subtracted from the bbp(700 nm) values derived above.

Colored dissolved organic matter; Acquired with WET Labs MCOMS CDOM Fluorometer (excitation 370 nm/emission 460 nm). A dark offset was subtracted from the raw sensor counts and the result was multiplied by a factory-determined scale factor to obtain colored dissolved organic matter concentration [ppb].

The dark offset was computed separately for each sensor as the mean of the surface minima (0 – 20 dbar) measured during all profiles between July and September 2013 at BATS (https://www.bco-dmo.org/dataset/728371).

Uncorrected particulate beam attenuation coefficient cp (650 nm); Acquired with WET Labs c-ROVER 2000 transmissometer with 650-nm wavelength and 0.25-m pathlength.

Transmittance is calculated as:

                Transmittance = (Signal - Dark) / (Cal Signal - Dark)
                Signal = raw output in counts
                Dark = counts with beam blocked, factory supplied
                Cal Signal = counts with Milli-Q water in sensor path, acquired prior to deployment (https://www.bco-dmo.org/dataset/728371)

The beam attenuation coefficient is calculated as:

                cp,uncorr(650 nm) = -ln (transmittance) / pathlength [m]

No correction for drift of the sensor over time (for instance, due to bio-fouling, see Estapa et al. 2013) has been applied.

The uncalibrated compass heading of the float.

Number of float park cycle.