Open AccessCorrection to Beaufort‐estimated wind speeds over the Tropical Indian Ocean
Author(s) -
Kameshwari Nunna,
Udaya Bhaskar T. V. S.,
Pattabhi Rama Rao E.,
Jampana Venkata
Publication year - 2018
Publication title -
meteorological applications
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.672
H-Index - 59
eISSN - 1469-8080
pISSN - 1350-4827
DOI - 10.1002/met.1731
Subject(s) - beaufort scale , anemometer , climatology , scale (ratio) , environmental science , meteorology , indian ocean , wind speed , geography , geology , oceanography , cartography
The Beaufort Scale was used to estimate wind speeds (WSs) over oceans before the introduction of ship‐mounted anemometers. Beaufort‐estimated WSs form a major component of most historical data sets such as those of the International Comprehensive Ocean—Atmosphere Dataset (ICOADS) and data obtained from the Indian Meteorological Department (IMD). The Beaufort‐estimated WSs in the individual records of the ICOADS and IMD data are based on the World Meteorological Organization (WMO) 1100 scale, which gives biased WS equivalents. Corrections to this scale have been formulated to give better estimates of the WS. Among others, the corrections proposed by Da Silva and Lindau are found to give least bias. However, the correction scale derived by Da Silva has limited utility over the Indian Ocean. The present paper derives a correction explicitly for the Tropical Indian Ocean (30–100 ° E, 30 ° S–30 ° N) using individual records from the ICOADS and IMD unique data for the period 1985–2005. There is significant improvement in the agreement between Beaufort estimates and the anemometer‐measured WSs based on these corrections. The regression co‐efficients for the new scale are derived using annual and monthly data. The correction scale derived, which is based on the regression co‐efficients for July (July scale), gives less bias compared with the scales of other months. The slope differs by almost 17% when the new July scale is applied to the annual data when compared with the Da Silva scale; the bias showed a reduction from 0.52 to 0.08 m/s.