Analysis of diurnal to seasonal variability of Integrated Water Vapour in the South Indian Ocean basin using ground‐based GNSS and fifth‐generation ECMWF reanalysis (ERA5) data

Abstract

Abstract The spatial and temporal distribution of tropospheric water vapour in the South Indian Ocean (SIO) basin is investigated using observations collected from twelve GNSS stations spanning the basin. The comparison of GNSS‐derived integrated water vapour (IWV) content against radiosoundings and satellite‐borne microwave radiometer data shows good agreement, with global uncertainties ranging from 0.76 to 1.17 kg · m −2 , depending on GNSS station locations. GNSS‐derived IWV contents show a strong seasonal cycle, characterized by higher water vapour content during the austral summer, when the InterTropical Convergence Zone (ITCZ) is located in the Southern Hemisphere. At the seasonal time‐scale, the observed annual IWV amplitude varies from 10 to 15 kg · m −2 near the Equator to 20 to 30 kg · m −2 in the Subtropics. The GNSS IWV signature of the Madden–Julian Oscillation (MJO) is hardly noticeable during the Austral winter, but varies from 1–2 to 4 kg · m −2 between the active and suppressed phases of the MJO during austral summer. At diurnal time‐scales, GNSS IWV shows larger diurnal amplitude over land (2–3 kg · m −2 ) than over open ocean (1–2 kg · m −2 ), with highest amplitudes (up to 7 kg · m −2 ) observed over large and mountainous islands. The phase analysis of the IWV diurnal cycle indicates that the diurnal maximum (minimum) is reached in the late afternoon/evening (morning) over land, at night (mid‐day) over ocean and in the early morning (late afternoon) at coastal locations. A comparison of GNSS‐derived IWV contents against fifth‐generation European Centre for Medium‐range Weather Forecasts (ECMWF) Reanalysis (ERA5) data shows that ERA5 generally correctly reproduces the IWV content at both seasonal, intra‐seasonal and diurnal time‐scales, although some discrepancies can be noticed over small islands characterised by steep orography. The signature of the MJO in ERA5 also shows good agreement with GNSS observations at most studied locations.