Spatiotemporal variation and mechanisms of temperature inversion in the Bay of Bengal and the eastern equatorial Indian Ocean

In the northern Bay of Bengal, the existence of intense temperature inversion during winter is a widely accepted phenomenon. However, occurrences of temperature inversion during other seasons and the spatial distribution within and adjacent to the Bay of Bengal are not well understood. In this study, a higher resolution spatiotemporal variation of temperature inversion and its mechanisms are examined with mixed layer heat and salt budget analysis utilizing long-term Argo（2004 to 2020） and RAMA（2...     

An assessment of TropFlux and NCEP air-sea fluxes on ROMS simulations over the Bay of Bengal region

This study presents an assessment of the TropFlux and the National Centers for Environmental Prediction (NCEP) reanalysis air-sea fluxes in simulating the surface and subsurface oceanic parameters over the Bay of Bengal (BoB) region during 2002–2014 using the Regional Ocean Modelling System (ROMS). The assessment has been made by comparing the simulated fields with in-situ and satellite observations. The simulated surface and subsurface temperatures in the TropFlux forced experiment (TropFlux-E) show better agreement with the Research Moored Array for African-Asian-Australian Monsoon Analysis (RAMA) and Argo observations than the NCEP forced experiment (NCEP-E). The BoB domain averaged sea surface temperature (SST) simulated in the NCEP-E is consistently cooler than the satellite SST, with a root mean square error (RMSE) of 0.79 °C. Moreover, NCEP-E shows a limitation in simulating the observed seasonal cycle of the SST due to substantial underestimation of the pre-monsoon SST peak. These limitations are mostly due to the lower values of the NCEP net heat flux. The seasonal and interannual variations of SST in the TropFlux-E are better comparable to the observations with correlations and skills more than 0.80 and 0.90 respectively. However, SST is overestimated during summer monsoon periods mainly due to higher net heat flux. The superiority of TropFlux forcing over the NCEP reanalysis can also be seen when simulating the interannual variabilities of the magnitude and vertical extent of Wyrtki jets at two equatorial RAMA buoy locations. The jet is weaker in the NCEP-E relative to the TropFlux-E and observations. The simulated sea surface height anomalies (SSHA) from both the experiments are able to capture the regions of positive and negative SSHA with respect to satellite-derived altimeter data with better performance in the TropFlux-E. The speed of the westward propagating Rossby wave along 18°N in the TropFlux-E is found to be about 4.7 cm/s, which is close to the theoretical phase speed of Rossby waves.     

Evaluation of GODAS Using RAMA Mooring Observations from the Indian Ocean

We present a comparison of the Global Ocean Data Assimilation System (GODAS) five-day ocean analyses against in situ daily data from Research Moored Array for African-Asian-Australian Monsoon Analysis and Prediction (RAMA) moorings at locations 90°E, 12°N; 90°E, 8°N; 90°E, 0°N and 90°E, 1.5°S in the equatorial Indian Ocean and the Bay of Bengal during 2002–2008. We find that the GODAS temperature analysis does not adequately capture a prominent signal of Indian Ocean dipole mode of 2006 seen in the mooring data, particularly at 90°E 0°N and 90°E 1.5°S in the eastern India Ocean. The analysis, using simple statistics such as bias and root-mean-square deviation, indicates that standard GODAS temperature has definite biases and significant differences with observations on both subseasonal and seasonal scales. Subsurface salinity has serious deficiencies as well, but this may not be surprising considering the poorly constrained fresh water forcing, and possible model deficiencies in subsurface vertical mixing. GODAS reanalysis needs improvement to make it more useful for study of climate variability and for creating ocean initial conditions for prediction.