16:30 〜 16:45
[AHW24-23] Estimation of freshwater discharge from the Gulf of Alaska drainage basins
キーワード：Hydrology of Alaska mountain basins, SWAT model, Glacier
Recent studies suggested an overturning circulation of the North Pacific seawater is very important to the biogeochemical environment in the North Pacific Ocean and potentially the global climate, as it transports cold, oxygenated and nutrient-rich water from the surface into deeper layers of the ocean. The strength of this circulation is mainly affected by surface seawater salinity along the circulation pathway, which includes regions such as the Alaskan Stream, the Bering Sea, and the Sea of Okhotsk. What may influence the surface salinity in such regions therefore became an important topic for a better understanding of the circulation. In this study, we considered the freshwater discharge into the Gulf of Alaska (GOA) as one of the main possibilities, as it is directly connected to the Alaska Stream region.
Five representative basins around the GOA were selected as the focus of interest. In these basins, glacier is one major feature, whose mass balance changes may affect river discharge at a non-negligible level, while to date, glacier mass balance was observed only over a few specific glaciers for a limited time period. For obtaining continuous time-series data of the glacierized area both within and outside the five focused basins, we estimated the glacier-melt water from 1983 to 2013 with a temperature-index (TI) method, which relies on the empirical relationship between air temperature and snow or ice melt. Estimated results of the glacier-melt water in the five focused basins were then treated as an additional input for a basin-scale hydrological model named Soil and Water Assessment Tool (SWAT). After proper calibrations, SWAT-simulated monthly discharge matches the observation reasonably well (NSE > 0.6, r2 > 0.8) and exhibits a strong seasonal variability but with no significant trend. From November to March, discharge remains at a base-flow level as river becomes relatively inactive due to substantially low temperature. Discharge increases sharply around April, as extensive snowmelt occurs when temperature rises above the freezing point. A dramatic increase of discharge ends shortly after temperature peaked and started decreasing around July, which suggests that snowmelt water likely contributes to discharge more dominantly than rainfall during summer season. Discharge keeps decreasing from July to November with occasional and minor fluctuation, then returns to base flow level before December. During 1983-2013, the estimated annual mean discharge from the five basins is around 5,933 ± 558 m3 ·s-1 with the contribution of 20.9% from the glacier melt. The discharge correlates negatively with the salinity anomaly of the Alaskan Stream (AS) by a five-month lag during 1985-2008 (p < 0.05, r2 = 0.19). Our result suggests that the freshwater discharge into the GOA may be connected to the overturning circulation by affecting surface seawater salinity along the circulation pathway.
Five representative basins around the GOA were selected as the focus of interest. In these basins, glacier is one major feature, whose mass balance changes may affect river discharge at a non-negligible level, while to date, glacier mass balance was observed only over a few specific glaciers for a limited time period. For obtaining continuous time-series data of the glacierized area both within and outside the five focused basins, we estimated the glacier-melt water from 1983 to 2013 with a temperature-index (TI) method, which relies on the empirical relationship between air temperature and snow or ice melt. Estimated results of the glacier-melt water in the five focused basins were then treated as an additional input for a basin-scale hydrological model named Soil and Water Assessment Tool (SWAT). After proper calibrations, SWAT-simulated monthly discharge matches the observation reasonably well (NSE > 0.6, r2 > 0.8) and exhibits a strong seasonal variability but with no significant trend. From November to March, discharge remains at a base-flow level as river becomes relatively inactive due to substantially low temperature. Discharge increases sharply around April, as extensive snowmelt occurs when temperature rises above the freezing point. A dramatic increase of discharge ends shortly after temperature peaked and started decreasing around July, which suggests that snowmelt water likely contributes to discharge more dominantly than rainfall during summer season. Discharge keeps decreasing from July to November with occasional and minor fluctuation, then returns to base flow level before December. During 1983-2013, the estimated annual mean discharge from the five basins is around 5,933 ± 558 m3 ·s-1 with the contribution of 20.9% from the glacier melt. The discharge correlates negatively with the salinity anomaly of the Alaskan Stream (AS) by a five-month lag during 1985-2008 (p < 0.05, r2 = 0.19). Our result suggests that the freshwater discharge into the GOA may be connected to the overturning circulation by affecting surface seawater salinity along the circulation pathway.