Questions about Ocean Sensors destruction:

[OKIsItJustMe]

9. Critical dependence of global ocean heat monitoring on the ocean observing system

Zhu, Y., Cheng, L., Trenberth, K.E. et al. Critical dependence of global ocean heat monitoring on the ocean observing system. Nat. Clim. Chang. (2026). https://doi.org/10.1038/s41558-026-02661-6

Critical dependence of global ocean heat monitoring on the ocean observing system

Received: 10 February 2026
Accepted: 29 April 2026
Published online: 22 May 2026

…

Ocean monitoring critically depends on the Global Ocean Observing System, which has provided upper-2,000-m temperatures with near-global coverage since around 2005 but is increasingly vulnerable to policy and economic pressures. Here we show that data reduction in the Global Ocean Observing System would substantially degrade its capability to monitor ocean heat content changes. We further highlight the shared responsibility of nations, through sustained international coordination and long-term national commitments, to maintain an adequate observing system.

…

To assess how the degradation of GOOS could affect climate monitoring, using OHC as an integrated metric, we conducted random data-removal experiments in which 20%, 40%, 60% and 80% of the existing GOOS observations were removed to mimic plausible observing-system degradation. Removal was performed separately for different instrument types, accounting for their distinct deployment and sampling characteristics (see Methods for the details). Annual rates of change in OHC (dOHC/dt), calculated as centred differences of annual mean OHC, were derived from the full GOOS dataset (used as a reference and hereafter denoted REF) and from each degraded configuration (Fig. 2a)

These experiments show a pronounced degradation in the ability to resolve year-to-year variations in the ocean heating rate as observational coverage decreases (Fig. 2a). The relative error in annual ocean heating rate increases by 33 ± 6%, 57 ± 6%, 79 ± 6% and 97 ± 20% for 20%, 40%, 60% and 80% data removal, respectively (Fig. 2b; uncertainties denote ±1 standard deviation across ten realizations). A relative error of 97% corresponds to a root-mean-square error (RMSE) of 0.34 W m⁻² in the global dOHC/dt time series, comparable to the actual interannual variability of dOHC/dt in REF (0.35 W m⁻²); this renders the global ocean heating signal difficult to distinguish from noise.

In addition to interannual variability, the observed increase in the rate of ocean warming implies that Earth’s heat accumulation is accelerating, with important implications for climate adaptation and mitigation15. Our experiments reveal that estimates of this accelera-tion are also impacted by the degradation of the observing system: acceleration biases reach 7 ± 4%, 11 ± 7%, 15 ± 10% and 17 ± 10% for 20%, 40%, 60% and 80% data removal, respectively (Fig. 2c). A 20% relative error corresponds to an RMSE of 0.1 W m⁻² dec⁻¹.

…