Systematic adjustments of hydrographic sections for internal consistency

Citation
Gc. Johnson et al., Systematic adjustments of hydrographic sections for internal consistency, J ATMOSP OC, 18(7), 2001, pp. 1234-1244
Citations number
11
Categorie Soggetti
Earth Sciences
Journal title
JOURNAL OF ATMOSPHERIC AND OCEANIC TECHNOLOGY
ISSN journal
07390572 → ACNP
Volume
18
Issue
7
Year of publication
2001
Pages
1234 - 1244
Database
ISI
SICI code
0739-0572(2001)18:7<1234:SAOHSF>2.0.ZU;2-T
Abstract
A significant legacy of the World Ocean Circulation Experiment (WOCE) is th e large number of high quality, high-resolution, full-depth, transoceanic h ydrographic sections occupied starting in the mid-1980s. The section data p rovide an unprecedented survey of World Ocean water properties. Most statio ns sampled pressure, temperature, salinity, dissolved oxygen, and nutrients (nitrate, phosphate, and silicic acid) at up to 36 depths. While the WOCE Hydrographic Program (WHP) strenuously advocated employing standardized mea surement techniques on all sections, small but significant systematic diffe rences among cruise legs are found. A simple method for adjusting measureme nts to maximize internal consistency is presented and applied to available WOCE data in the Pacific Basin. First, the sections are broken into distinc t cruise legs between port stops. Then, crossovers are identified where two different cruise legs cross or approach each other. Using hydrographic dat a from each cruise leg near each crossover, linear fits are made of propert ies on potential temperature surfaces against distance along cruise track. These fits are then used to evaluate property differences and their uncerta inties at crossovers. A set of least squares models are used to generate se ts of adjustments, with related uncertainties, for all the properties of ea ch cruise leg. These adjustments minimize differences of water properties a mong cruise legs at the crossovers in a least squares sense. The adjustment s can be weighted by difference uncertainties, and damped by a priori estim ates of the expected differences. Initial standard deviations of crossover differences are 0.0028 for salinity, 2.1% for oxygen, 2.8% for nitrate, 1.6 % for phosphate, and 2.1% for silicic acid. The adjustments roughly halve t hese values, bringing cruise legs into agreement within WHP target accuraci es.