3:30 PM - 5:00 PM
[AAS06-P07] Improvement of J-OFURO3 air-sea heat flux by correcting for near-surface air temperature bias in atmospheric reanalysis
Keywords:atmospheric reanalysis, air temperature, surface heat budget, satellite remote sensing
Atmospheric reanalysis data are from numerical simulations based on atmospheric general circulation models synthesized with observed data using data assimilation methods. Generally, it is provided as gridded data with no missing value and has been used for various research and applications. The third generation product from Japanese Ocean Flux Data Sets with Use of Remote Sensing Observations (J-OFURO3), which estimates air-sea surface heat flux based on satellite observations, also uses the NOAA-provided atmospheric reanalysis data (NCEP) as the near-surface air temperature value over the global ocean. However, the near-surface air temperature data in NCEP is known to have a positive bias, which is pointed out to be the cause of the bias in the global heat budget in J-OFURO3.
This study aims to investigate the characteristics of the bias in NCEP near-surface air temperature data through comparison with in-situ observations and to propose an effective correction method for the bias. The obtained correction method is then applied to estimating global ocean near-surface air temperature data and surface sensible heat fluxes to investigate its effectiveness. In addition, the cause of the bias in near-surface air temperature in NCEP was discussed.
Using in-situ buoy observation data to investigate the characteristics of NCEP's near-surface air temperature bias, we found that the bias is strongly related to changes in air temperature itself. Specifically, we revealed a tendency for positive bias to occur when air temperatures are lower. This is consistent with the existence of a latitude-dependent bias pointed out in previous studies, but it also indicates that the bias is characterized by a bias closer to the actual situation. Therefore, we developed an empirical equation to correct the bias in NCEP's near-surface air temperature data based on in-situ global buoy observations. As a result of the verification using data independent of the data used to develop the empirical equation, we succeeded in improving most of the biases that had occurred in the global region. In particular, it was shown that the positive bias of about +0.75 °C in the temperature range of air temperatures below 15 °C could be reduced to +0.03 °C by the correction. Furthermore, the correction of the global near-surface air temperature bias also improves the value of the surface sensible heat flux in J-OFURO3, indicating that the correction is effective in reducing the bias in the global heat budget, which has been pointed out in previous studies.
Regarding the cause of the near-surface air temperature bias, focusing on a time scale of several days, the temperature bias corresponds well with the change in the surface sensible heat flux, and the NCEP sensible heat flux is larger than that estimated from in-situ observations, suggesting that the sensible heat flux bias is related to the temperature bias in the NCEP atmospheric reanalysis system.
The NCEP atmospheric reanalysis data is one of the data sets that has been analyzed and provided data since its publication in 1996 to the present. Similar temperature bias characteristics are also found in other, more recent-generation atmospheric reanalysis data. The results of this study provide useful information for general atmospheric reanalysis users and contribute to the future development of numerical models, data assimilation, and atmospheric reanalysis data. In addition, they will contribute to providing better fundamental data for the observational estimation of the global surface heat flux, such as J-OFURO3.
This study aims to investigate the characteristics of the bias in NCEP near-surface air temperature data through comparison with in-situ observations and to propose an effective correction method for the bias. The obtained correction method is then applied to estimating global ocean near-surface air temperature data and surface sensible heat fluxes to investigate its effectiveness. In addition, the cause of the bias in near-surface air temperature in NCEP was discussed.
Using in-situ buoy observation data to investigate the characteristics of NCEP's near-surface air temperature bias, we found that the bias is strongly related to changes in air temperature itself. Specifically, we revealed a tendency for positive bias to occur when air temperatures are lower. This is consistent with the existence of a latitude-dependent bias pointed out in previous studies, but it also indicates that the bias is characterized by a bias closer to the actual situation. Therefore, we developed an empirical equation to correct the bias in NCEP's near-surface air temperature data based on in-situ global buoy observations. As a result of the verification using data independent of the data used to develop the empirical equation, we succeeded in improving most of the biases that had occurred in the global region. In particular, it was shown that the positive bias of about +0.75 °C in the temperature range of air temperatures below 15 °C could be reduced to +0.03 °C by the correction. Furthermore, the correction of the global near-surface air temperature bias also improves the value of the surface sensible heat flux in J-OFURO3, indicating that the correction is effective in reducing the bias in the global heat budget, which has been pointed out in previous studies.
Regarding the cause of the near-surface air temperature bias, focusing on a time scale of several days, the temperature bias corresponds well with the change in the surface sensible heat flux, and the NCEP sensible heat flux is larger than that estimated from in-situ observations, suggesting that the sensible heat flux bias is related to the temperature bias in the NCEP atmospheric reanalysis system.
The NCEP atmospheric reanalysis data is one of the data sets that has been analyzed and provided data since its publication in 1996 to the present. Similar temperature bias characteristics are also found in other, more recent-generation atmospheric reanalysis data. The results of this study provide useful information for general atmospheric reanalysis users and contribute to the future development of numerical models, data assimilation, and atmospheric reanalysis data. In addition, they will contribute to providing better fundamental data for the observational estimation of the global surface heat flux, such as J-OFURO3.