Profiling Float Data

A profiling float (#12592; APEX, Webb Research Inc., Falmouth, USA) equipped with an oxygen Optode sensor (Model 4330, Aanderaa), an SBE‐41 CTD instrument (Sea‐Bird Scientific), as well as a chl fluorometer and volume scattering coefficient (at 140 degrees, 700 nm) meter (ECO FLBB, WetLABS) was deployed at ~42N/158W on April 18, 2019. The mission design consisted of ascension from 1000‐m depth to surface twice a day, at ~local noon and local midnight. The float provided its last profile on October 25, 2019, when batteries were exhausted. At each surfacing event, the optode collected measurements in air (~ 25 cm above sea surface) to allow post-calibration. Raw float oxygen optode data were corrected for pressure and salinity following Uchida et al. [2008] and Garcia and Gordon [1992]. Float optode oxygen data were calibrated using optode air measurements taken at the time of each float surfacing following methods described in Bushinsky et al. [2016]. Briefly, air calibration relies on the estimate of a gain factor G, such that O2_corrected = G×O2_raw. In practice, G is an average gain factor for each float profile determined from the ratio of the expected partial pressure of oxygen in air (pO2) to the partial pressure of oxygen in air measured by the optode (pO2_optode), i.e. gi=pO2/pO2_optode. Optode air measurements were first filtered to remove outliers (e.g. measurements taken underwater or with high variance), and mean values per surfacing were recorded. pO2_optode was calculated from optode phase and temperature as in the Aanderaa Manual. Atmospheric pO2 was calculated as in Bushinsky et al. [2016] using float-derived water vapor pressure estimates (pH20) at the time of each float surfacing (Aanderaa Manual) as well as the 6-h NOAA NCEP atmospheric pressure and surface relative humidity data interpolated to the time and location of each float surfacing. An average G value of 1.30 was then calculated based on the gi estimates from each surfacing at each station, and applied to all optode profiles. Gain-corrected oxygen concentration and saturation data were not calibrated against Winkler measurements, and therefore variables are noted as being "uncalibrated". Potential density was calculated using the Gibbs gsw_sigma0.m function in Matlab. Chlorophyll fluorescence was converted to chlorophyll concentrations by first subtracting a deep-value (average at ~980m depth) and applying the factory scaling factor. The volume scattering coefficient was calibrated by subtracting a dark value obtained on ship deck prior to deployment, and applying the factory scaling factor. Volume scattering coefficient was converted to particulate backscattering coefficient (bbp, the product provided in this dataset) using the approach of Zhang et al [2009] to obtain backscattering due to seawater, and then by using the approach of Boss and Pegau [2001] to obtain bbp, assuming a X factor of 1.1 (see Boss and Pegau [2001]). References cited:Uchida, H., Kawano, T., Kaneko, I., & Fukasawa, M. (2008). In situ calibration of Optode‐based oxygen sensors. Journal of Atmospheric and Oceanic Technology, 25(12), 2271–2281. //// García, H. E., & Gordon, L. I. (1992). Oxygen solubility in seawater: Better fitting equations. Limnology and Oceanography, 37(6), 1307–1312. //// Bushinsky, S. M., Emerson, S. R., Riser, S. C., & Swift, D. D. (2016). Accurate oxygen measurements on modified argo floats using in situ air calibrations. Limnology and Oceanography: Methods, 14(8), 491–505. //// Boss, E. S., & Pegau, W. S. (2001). Relationship of light scattering at an angle in the backward direction to the backscattering coefficient. Applied Optics, 40, 5503–5507. //// Zhang, X., Hu, L., &He, M-X (2009). Scattering by pure seawater at high salinity. Optics Express, 17(15), 12685. https://doi.org/10.1364/oe.17.012685

剖面浮标（profiling float）#12592搭载APEX型平台（Webb Research Inc.，美国法尔茅斯），配备氧气光极传感器（Optode，型号4330，Aanderaa公司）、SBE-41型CTD（温盐深剖面仪）（Sea-Bird Scientific公司），以及叶绿素荧光计（chlorophyll fluorometer）与700nm、140°体积散射系数（volume scattering coefficient）测量仪（ECO FLBB，WetLABS公司），于2019年4月18日部署于约42°N/158°W海域。
该浮标的观测任务设计为每日两次从1000米深度上浮至海面，分别在当地正午与当地午夜时段执行。
该浮标于2019年10月25日电池耗尽时，完成了最后一次剖面观测。
每次上浮过程中，光极传感器会在海面上方约25厘米处采集空气数据，用于后续校准。
浮标原始氧气光极数据依据Uchida等人（2008）与Garcia及Gordon（1992）提出的方法，进行了压力与盐度校正。
浮标氧气光极数据依据Bushinsky等人（2016）描述的方法，利用每次浮标上浮时采集的空气光极测量数据完成校准。
简言之，空气校准依赖于增益因子G的估算，校正后氧气浓度满足公式：O2_corrected = G × O2_raw。实际应用中，G为每个浮标剖面的平均增益因子，由空气中理论氧气分压（pO2）与光极传感器测得的空气中氧气分压（pO2_optode）的比值确定，即gi = pO2 / pO2_optode。
首先对光极空气测量数据进行滤波以剔除异常值（例如水下采集或高方差的测量数据），并记录每次上浮的平均测量值。
依据Aanderaa仪器手册的方法，由光极相位与温度计算得到pO2_optode。
大气氧气分压pO2的计算依据Bushinsky等人（2016）的方法，使用每次浮标上浮时由浮标估算得到的水汽压（pH₂O），以及NOAA NCEP的6小时分辨率大气压强与地表相对湿度数据，插值至每次浮标上浮的时间与位置后参与计算。
随后基于各站位每次上浮得到的gi估算值，计算得到平均增益因子G为1.30，并将其应用于所有光极传感器剖面数据。
经增益校正后的氧气浓度与饱和度数据未通过Winkler滴定法进行校准，因此相关变量标注为"uncalibrated"。
位势密度的计算采用Matlab中的Gibbs gsw_sigma0.m函数完成。
叶绿素荧光数据转换为叶绿素浓度的步骤为：首先扣除深部基准值（约980米深度处的平均荧光值），随后应用出厂标定的缩放因子。
体积散射系数的校准步骤为：扣除部署前在船甲板上测得的暗电流值，随后应用出厂标定的缩放因子。
体积散射系数首先依据Zhang等人（2009）的方法转换得到海水本底后向散射系数，随后依据Boss与Pegau（2001）的方法，假设X因子为1.1（详见Boss及Pegau，2001），转换得到颗粒后向散射系数（particulate backscattering coefficient，即本数据集提供的产物bbp）。
引用文献：
Uchida, H., Kawano, T., Kaneko, I., & Fukasawa, M. (2008). 内田浩一, 川野智, 金子一史, 深泽正明. (2008). 基于光极的氧气传感器原位校准. 《大气与海洋技术学报》, 25(12), 2271–2281.
García, H. E., & Gordon, L. I. (1992). 加西亚H.E., 戈登L.I. (1992). 海水中的氧气溶解度：更优的拟合公式. 《湖沼学与海洋学》, 37(6), 1307–1312.
Bushinsky, S. M., Emerson, S. R., Riser, S. C., & Swift, D. D. (2016). 布什斯基S.M., 爱默生S.R., 赖泽S.C., 斯威夫特D.D. (2016). 基于原位空气校准的改装Argo浮标高精度氧气测量. 《湖沼学与海洋学：方法》, 14(8), 491–505.
Boss, E. S., & Pegau, W. S. (2001). 博斯E.S., 佩高W.S. (2001). 后向角度光散射与后向散射系数的关系. 《应用光学》, 40, 5503–5507.
Zhang, X., Hu, L., & He, M-X (2009). 张X, 胡L, 何M-X (2009). 高盐度纯海水的散射特性. 《光学快报》, 17(15), 12685. https://doi.org/10.1364/oe.17.012685