Reducing Temperature Biases in Tropical Ocean Models

  • 24 November 2020
Reducing Temperature Biases in Tropical Ocean Models

A key aspect of ocean models is accurately representing sea surface temperatures. There are certain regions of the ocean, however, where models commonly have difficulty in correctly reproducing recorded temperatures. Two of these regions are an area in the eastern equatorial Pacific, which models commonly describe as cooler than observed, and a region off the coast of South America, which models commonly describe as warmer than observed. The mixing of warm and cool water between ocean layers plays an important role in controlling the temperature of the upper ocean. Thus, a better representation of the temperature gradient, or thermocline, in the upper ocean could reduce these common model errors, or biases. Researchers from the University of Hawai’i at Mānoa compare two model experiments to explore how increasing the vertical detail in an ocean model impacts the modeled sea surface temperature in the tropical Pacific Ocean. Their work was funded by CPO’s Climate Variability & Predictability (CVP) program as part of NOAA’s contributions to the Tropical Pacific Observing System (TPOS) process studies. The only difference between the two models used in the study is that one model has a much finer representation or resolution of small-scale vertical features in the upper ocean. Their findings, published in Ocean Modelling, show that the high-resolution model seems to reduce the commonly observed biases in the two regions, both warming the eastern Pacific and cooling offshore of South America, respectively. The researchers explain the probable processes at play, explaining step by step the changes in the model that carry through from improved resolution at the start to reduced sea surface temperature biases at the finish. These processes are different for different regions: sometimes shear-generated turbulence is the primary contributor and sometimes it is ocean stratification or layering. Regardless, the study recommends that ocean models incorporate finer vertical resolution in order to capture small-scale vertical features that impact ocean mixing and thus influence larger-scale ocean structure. The researchers also note that they only showed improvements based on fine-tuning ocean models. Coupling their experiments with atmospheric models could show even larger improvements in fixing sea surface temperature biases. 

Read the study »




1315 East-West Highway Suite 100
Silver Spring, MD 20910


Americans’ health, security and economic wellbeing are tied to climate and weather. Every day, we see communities grappling with environmental challenges due to unusual or extreme events related to climate and weather.