A Multi-Inventory Ensemble Analysis Of The Effects Of Atmospheric Rivers On Precipitation And Streamflow In The Namgang-Dam Basin In Korea

Atmospheric rivers (ARs) are an important concern in regional water management; however, little is known about the AR impacts on hydrology in East Asia (EA). This study analyzes the characteristics of storms, precipitation (P), streamflow (Q), and runoff coefficient (R) in the Namgang-dam basin in Korea related to ARs for 2000-2013, as well as the sensitivity of the analysis results to AR detection methods, using observed P, Q, and three different AR inventories. The basin experiences 37.3 storms annually, of which 54% are AR storms that provide over 60% of the annual P and Q. The AR (non-AR) storms are dominant in storm frequency and the storm-total P and Q for January-July (August-December) with peaks in July (August). The monthly AR frequency varies closely with the seasonal variations in the EA monsoon and the North Pacific storm track which modulate the number of extratropical cyclones. The AR storms produce most of the extreme events; they also generate larger storm-mean P and Q than the non-AR storms for all months. The seasonal variations in R are related to the total (AR- and non-AR storms combined) P through the seasonal soil water variations, making the AR effects on R unclear. Considering 95% confidence intervals, the AR storms are distinguished well from the non-AR storms in storm frequency and the storm-total and storm-mean P and Q. The sensitivity to AR inventories is not critical in quantifying the AR-storm characteristics and their impacts on hydrologic variables except for R.