| Summary: | This thesis investigates some important aspects of the atmospheric branch of the hydrological cycle in the modern day climate from an observational perspective. Data quality is evaluated, focusing on two state-of-the-art reanalysis products, ERA-I and JRA-55. Regional-scale discrepancies among reanalyses and observations, especially in their annual cycles, are found in the warm pool, Amazon, Gulf stream and Indian subcontinent regions. In the tropics, oceanic evaporation and its temporal variability are notably greater in JRA-55 than in ERA-I and satellite-based estimates, while both reanalyses overestimate precipitation. Higher tropical precipitation and evaporation, accompanied by a slightly lower level of total column water (TCW), might suggest a more intense hydrological cycle, but this can be an ill-defined concept especially when analysis increments mask “spin-down” errors in reanalysis models. Analysis increments arise to remove unphysical residuals in the atmospheric water budget, and these are explored via a cluster analysis to identify regimes with common behavior. Consistent for ERA-I and JRA-55, the regime with the largest negative residuals (greater moisture outputs than inputs) exceeding 50% of mean precipitation occurs during the dry season of some low latitude regions that feature strong seasonality, high evapotranspiration and high moisture divergence. Errors in the moisture divergence are likely responsible because they correlate strongly with the budget residual. Empirical Orthogonal Function (EOF) and Self Organizing Map (SOM) analyses are applied to identify the dominant inter-annual patterns of vertically-integrated moisture divergence variability. They reveal that the transition from strong La Niña through to extreme El Niño events is not a linear one and that the EOF orthogonality constraint results in the patterns being split between leading EOFs that are non-linearly related. The SOM analysis captures the range of responses to the El Niño Southern Oscillation (ENSO), indicating that the distinction between the moderate and extreme El Niños can be as great as the difference between La Niña and moderate El Niños, from a moisture divergence point of view. On diurnal time scales, horizontal moisture fluxes vary in response to thermodynamic and dynamic effects. TCW shows a global scale diurnal cycle that peaks around 1800 - 2100 local time with a peak-to-trough magnitude of 0.4mm. Semi-diurnal variations in surface winds and pressure, consistent with atmospheric tidal theory, create a westward propagating moisture convergence/divergence wave along the equator. Finally, the importance of Tropical Cyclones (TCs) as a source of freshwater for the North American continent is estimated using an ensemble of schemes designed to attribute onshore moisture fluxes to TCs. Averaged over the 2004–2012 hurricane seasons and integrated over the western, southern and eastern coasts of North America, the seven schemes attribute 7 to 18% (mean 14 %) of total net onshore flux to Atlantic TCs. A reduced contribution of 10% (range 9 to 11 %) was found for the 1980–2003 period, though only two schemes could be applied to this earlier period. Over the whole 1980–2012 period, a further 8% (range 6 to 9% from two schemes) was attributed to East Pacific TCs, resulting in a total TC contribution of 19% (range 17 to 22 %) to the ocean-to-land moisture transport onto the North American continent between May and November. The inter-annual variability does not appear to be strongly related to ENSO.
|
|---|
