The Copernicus Surface Velocity Platform drifter with Barometer and Reference Sensor for Temperature (SVP-BRST): genesis, design, and initial results

Abstract

Abstract. To support calibration and validation of satellite sea surface temperature (SST) retrievals, over 60 high-resolution SST (HRSST) drifting buoys were deployed at sea between 2012 and 2017. Their data record is reviewed here. It is confirmed that sea state and immersion depth play an important role in understanding the data collected by such buoys and that the SST sensors need adequate insulation. In addition, calibration verification of three recovered drifters suggests that the sensor drift is low, albeit negative at around −0.01 K year−1. However, the statistical significance of these results is limited, and the calibration procedure could not be exactly reproduced, introducing additional uncertainties into this drift assessment. Based on lessons learnt from these initial buoys, a new sensor package for the Surface Velocity Platform with Barometer (SVP-B) was designed to serve calibration of SST retrievals by European Union's Copernicus satellites. The novel sensor package includes an HRSST sensor calibrated by a metrology laboratory. The sensor includes a pressure probe to monitor immersion depth in calm water and acquires SST data at 1 Hz over a 5 min interval every hour. This enables the derivation of mean SST as well as several percentiles of the SST distribution. The HRSST sensor is calibrated with an uncertainty better than 0.01 K. Analysis of the data collected by two prototypes deployed in the Mediterranean Sea shows that the buoys are able to capture small-scale SST variations. These variations are found to be smaller when the sea state is well mixed and when the buoys are located within eddy cores. This affects the drifter SST data representativeness, which is an aspect of importance for optimal use of these data.

Observational Needs of Sea Surface Temperature

“…Continued studies and activities are needed on the inter-comparison of FRM (Donlon et al, 2015) with data; the estimation and provision of the measurement uncertainties; and ensuring SI-traceability is established through international collaboration. In an effort to improve the sampling of SI-traceable in situ SSTs, a new generation of drifting buoys is being designed and tested through efforts of GHRSST, DBCP, European Union's Copernicus Programme and EUMETSAT (Poli et al, 2019). These efforts to improve in situ records are continuing and require considerable resources to improve the in situ SSTs not only as a standalone CDR but also as a verification tool for satellite-derived SST fields.…”

A dataset of hourly sea surface temperature from drifting buoys

“…What our uncertainty estimates are not able to capture is any drifter-specific original bias of a SST sensor, and which may, or not, have evolved in time since the times of manufacture and deployment (i.e. a sensor drift) 34 .…”

Harmonization of Space-Borne Infra-Red Sensors Measuring Sea Surface Temperature

“…The most obvious alternative would be SSTs measured in situ by drifting buoys. Although individual drifting buoys are not (for the most part) specifically calibrated or SI-traceable (with some exceptions [36]), it seems reasonable to assume that the multiple sensors across the drifting buoy network have a low average bias (because of compensating independent errors). However, drifting buoys measure the SST at a point location and at a depth of tens of centimeters (the exact depth varying with the drifting buoy design and sea state); they do not directly measure the skin SST over a pixel of ≥1 km 2 , to which MTA is actually sensitive.…”