Land Breeze Characteristics Off the West Coast of Sumatra

Properties of the land breeze off the west coast of Sumatra are explored using hourly 10-m winds from ERA5 for 2018. Data are analyzed for two sites--Padang and Bengkulu--with an emphasis on the frequency, timing, duration, maximum wind speed, and seasonality of the land breeze. The land breeze signal was isolated by first transforming the winds to coast parallel and coast perpendicular components, and then passing the shore perpendicular component through a combination of high-pass and low-pass filters, such that the filtered winds are representative of the diurnal circulation. A land breeze event was identified if a wind reversal from onshore to offshore occurred and the wind remained offshore for more than three hours. Of the 365 days analyzed, a land breeze was detected at Padang for all but two of the days. The Padang land breeze events had an average start time of 21 LST, a duration of roughly 11 hours, and a median maximum wind speed of 1.52 m/s that occurred during the early morning hours, near 5 LST. Similar land breeze characteristics were observed for the Bengkulu site.

About 30% of the land breeze events at each site exhibited an interruption of the land breeze, with winds reversing and becoming onshore briefly, before returning to offshore. On average, the interrupted land breeze events featured earlier onset times (~18 LST) and slower maximum wind speeds (<1 m/s), with the maximum offshore winds occurring during the evening hours (roughly 22 LST). Composites of the synoptic winds suggest that the interrupted land breeze events occur in environments with weaker onshore synoptic flow compared to the non-interruption events, which presumably accounts for the earlier onset time. At Bengkulu, the interrupted part of the land breeze features its maximum onshore flow at roughly 3 LST, suggesting a connection to the atmospheric semidiurnal solar tide (which is maximized near 3:30 LST). The maximum onshore winds during the interruption events at Padang occurred somewhat earlier, likely due to the influence of a large offshore island (Siberut Island) that has its own diurnal cycle and land breeze. In an extension of this work, observational data from the airport meteorological stations at Padang and Bengkulu will also be analyzed, allowing for a comparison between the ERA5 reanalysis and in-situ observations. The land breeze analysis will also be compared to a climatology of offshore rain events observed by the BMKG ground radar at Padang, to identify the relationship between the land breeze circulation and nocturnal propagating rainfall events.