Assessing Deep and Abyssal Ocean Heat Content Changes With a Dynamically Consistent Ocean State Estimate

Abstract Because of the sparseness of existing observations, our understanding of deep (2,000–4,000 m) and abyssal (>4,000 m) ocean heat content (OHC) changes remains limited. Previous studies utilizing repeated hydrographic section measurements identified a global warming trend in these layers. However, studies based on a widely used ocean state estimate ECCO v4 (Estimating the Circulation and Climate of the Ocean, version 4) show a contradictory cooling trend in the deep and abyssal Pacific and Indian Oceans. To examine if the sparseness of hydrographic measurements results in this contrasting conclusion, we conducted a sampling experiment with ECCO v4. Our results show that the signs of the OHC trends in the deep and abyssal oceans from the full spatial‐temporal data and the sampled data are generally consistent. The largest uncertainties mainly occur in regions where the deep ocean is dominated by newly formed deep‐water masses or where hydrographic sections are extremely sparse, such as the Northwest Atlantic Ocean and the Southern Ocean. Our findings also indicate that the discrepancies between ECCO v4 and observations in deep and abyssal OHC changes are not likely a sampling issue, and further studies are needed to determine the reasons.