Ocean state forecasting during VSCS Ockhi and a note on what we learned from its characteristics: A forecasting perspective

Abstract

Tropical Cyclone Ockhi was an intense cyclone, with a peculiar and long track, in the Arabian Sea in 2017. It caused severe damage to coastal infrastructure and death of 282 people. Indian National Centre for Ocean Information Services (INCOIS) issued the Joint INCOIS-IMD (India Meteorological Department) bulletins on the Ocean State Forecasts (OSF) and alerts/warnings during Ockhi. Validation of the OSF from INCOIS using buoys reveals that the forecasts were in good agreement with the observations [average correlation 0.9, RMSE <0.8 m (for larger waves), and scatter index <25%]. Climatological analysis of Genesis Potential Index (GPI) suggests that the southeast Arabian Sea, where the TC-Ockhi was intensified, had all the favourable conditions for intensification during November/December. Moreover, it was found that four days before the genesis of Ockhi, the environmental vorticity and relative humidity were more favourable for the cyclogenesis compared to vertical wind shear and potential intensity. The intensification rate was rapid as experienced by earlier cyclones in this region. Also, the cyclone track closely matched the background tropospheric winds. The present study suggests that the forecasters should look into the background dynamic and thermodynamic conditions extensively in addition to multi-model guidance to better predict the genesis, intensity and track of the cyclones.

Research highlights

• In the Arabian Sea, during the TC-Ockhi, the forecasts of wave parameters from the model forced with bias-corrected ECMWF winds resulted in very good agreement with observations.

• Climatologically, TC-Ockhi region has large potential for the genesis and intensification of TC due to an enhanced low-level cyclonic vorticity and the reduction in vertical wind shear.

• During the TC-Ockhi period, low-level vorticity and mid-tropospheric relative humidity were the dominant contributing factors, which lead to an enhanced GPI in the Arabian Sea.

• TC-Ockhi also had rapid intensification in a similar fashion the earlier cyclones in this region behaved.

• There is no abnormality also in the TC-Ockhi track, as the TC-Ockhi track matches well with the background tropospheric flow.

Appendix

Motivated from the work of Gray (1979), Emanuel and Nolan (2004) developed an empirical index named the Genesis Potential Index (GPI) to quantitatively describe the influence of large-scale environmental features on the TC genesis. Following, Emanuel and Nolan (2004), GPI can be represented as follows:

$$ {\text{GPI}} = \left( \frac{H}{50} \right)^{3}\,\times \left|10^{5} \eta\right|^{\frac{3}{2}} \times \left( {1 + 0.1{{V}}_{\text{shear}} } \right)^{ - 2} \times \left( {\frac{{V_{\text{pot}} }}{70}} \right)^{3}, $$

(1)

where H is the relative humidity (%) at 600 hPa, η is the absolute vorticity at 850 hPa (s⁻¹), V_shear is the magnitude of the vertical wind shear (m s⁻¹) between 850 and 200 hPa, and V_pot is the maximum TC potential intensity (PI) (m s⁻¹) defined by Emanuel (1999 and reference therein). The normalizing factors make the GPI dimensionless and the values were selected to keep the individual terms within the same order of magnitude.

For ease computation of relative roles of large-scale environmental factors to the total changes in the GPI, Li et al. (2013) modified the GPI equation as:

$$ \delta {\text{GPI}} = \alpha_1 \times \delta \left( \frac{H}{50} \right)^{3} +\, \alpha_2 \times \delta| 10^{5} \eta|^{\frac{3}{2}} + \alpha_3 \times \delta \left( {1 + 0.1{{V}}_{\text{shear}} } \right)^{ - 2} + \,\alpha_4 \times \delta \left( {\frac{{V_{\text{pot}} }}{70}} \right)^{3}, $$

(2)

where

$$ \begin{aligned} \alpha_1 & = \overline{{\left| {10^{5} \eta } \right|^{\frac{3}{2}} }} \times \overline{{\left( {1 + 0.1{{V}}_{\text{shear}} } \right)^{ - 2} }} \times \overline{{\left( {\frac{{V_{\text{pot}} }}{70}} \right)^{3} }} , \\ \alpha_2 & = \overline{{\left( \frac{H}{50} \right)^{3} }} \times \overline{{\left( {1 + 0.1{{V}}_{\text{shear}} } \right)^{ - 2} }} \times \overline{{\left( {\frac{{V_{\text{pot}} }}{70}} \right)^{3} }} , \\ \alpha_3 & = \overline{{\left( \frac{H}{50} \right)^{3} }} \times \overline{{\left| {10^{5} \eta } \right|^{\frac{3}{2}} }} \times \overline{{\left( {\frac{{V_{\text{pot}} }}{70}} \right)^{3} }} \\ \end{aligned} $$

and

$$ \alpha_4 = \overline{{\left( \frac{H}{50} \right)^{3} }} \times \overline{{\left| {10^{5} \eta } \right|^{\frac{3}{2}} }} \times \overline{{\left( {1 + 0.1{{V}}_{\text{shear}} } \right)^{ - 2} }} . $$

In the above expressions, the horizontal bar indicates mean state (climatology) and δ represents the perturbation (anomaly) of the individual parameter.