Vol.:(0123456789)
Natural Hazards (2020) 102:557–574
https://doi.org/10.1007/s11069-019-03756-4
1 3
ORIGINAL PAPER
Inter‑comparison of model, satellite and in situ tropical
cyclone heat potential in the North Indian Ocean
Babita Jangir
1
· D. Swain
1
· Samar Kumar Ghose
1
· Rishav Goyal
1
·
T. V. S. Udaya Bhaskar
2
Received: 19 April 2018 / Accepted: 14 August 2019 / Published online: 22 August 2019
© Springer Nature B.V. 2019
Abstract
The North Indian Ocean (NIO) experiences frequent tropical cyclones (TCs). TC heat
potential (TCHP) is a major ocean parameter responsible for TC genesis and intensifca-
tion changes. In this study, Indian National Centre for Ocean Information Services-Global
Ocean Data Assimilation System (INCOIS-GODAS) model and satellite-derived TCHP
data from National Remote Sensing Centre (NRSC) and National Oceanic and Atmos-
pheric Administration (NOAA) are validated against TCHP from in situ profles in the NIO
during the period 2011–2013 for buoys and during 2005–2015 for Argo data. Data from
eight moored buoys (6 in Bay of Bengal and 2 in Arabian Sea) under the Ocean Moored
Buoy Network are used. Comparison of model and in situ TCHP yields correlation coef-
cients (root-mean-square errors in kJ/cm
2
) of 0.74 (17.75), 0.59 (15.34), 0.70 (17.68), 0.60
(22.24), 0.57 (19.52), 0.73 (17.88) and 0.77 (39.17) at buoy locations BD08, BD09, BD10,
BD11, BD13, AD06 and AD10. The scatter indices between collocated TCHP values at
these locations were 0.32, 0.22, 0.30, 0.30, 0.31, 0.58 and 0.41. Further, it was found that
satellite-based TCHP from NRSC match better with in situ as compared to near-real-time
TCHP data obtained from NOAA. TCHP from INCOIS-GODAS model, NOAA delayed
time data and NRSC TCHP data set are also in good agreement with those from Argo
profles. As a case study, model and in situ TCHP were compared during a TC, “Thane” at
two buoy locations (BD11 and BD13), closest to its track. The analysis revealed underes-
timation of model TCHP at BD11, but good correlation at BD13. This could be attributed
to the existence of a strong temperature inversion at BD11. It is observed that although
the model is able to capture features like barrier and inversion layers, the temperature and
depth of such layers are underestimated. Further, the recovery time from the infuence of
TC on the ocean subsurface is also much longer in case of the model which thus needs to
be fne-tuned. Seasonal comparison of TCHP from various sources with in situ estimated
TCHP also shows better correlation between all the products for the pre-summer monsoon
compared to the post-summer monsoon season.
Keywords Tropical cyclone heat potential · OMNI buoy · Tropical cyclone · Cyclone
Thane
* Babita Jangir
bj11@iitbbs.ac.in
Extended author information available on the last page of the article