Long-term (> 40 Years) Monitoring of Near-Surface Salinity Needed to Detect Global Changes in the Hydrological Cycle
[16-Dec-2016] Reagan, J.R., Seidov, D., Boyer, T., Zweng, M.
Presented at the 2016 AGU Fall Meeting
Multiple studies have shown that since the mid-20th century near-surface salinity patterns have amplified, with fresh regions becoming fresher and salty regions becoming saltier. This pattern amplification is directly related to an amplification of the global hydrological cycle, with wet regions becoming wetter and dry regions becoming dryer. An amplified hydrological cycle could cause an increase in the number of extreme weather events, with more severe floods and droughts. However, it could also lead to additional freshwater fluxes over the deep water formation regions in the northern North Atlantic which can slow down the Atlantic Meridional Overturning Circulation. With historically unreliable global evaporation and precipitation data, salinity is the main source for tracking changes in the hydrological cycle. It has recently been shown that at time scales greater than 50-years, salinity pattern amplifications are significant. The question remains whether these significant amplification patterns can be seen at shorter time scales as well. To answer this question we performed near-surface salinity pattern amplification analyses to the near-global ocean for 10, 20, 30, 40, 50, and 60-year time periods utilizing pentadal (5-year) salinity anomalies derived from the World Ocean Database. Evaporation and precipitation (E-P) pattern amplifications are also computed over the more reliable data of the satellite era (last ~30 years) and are correlated with salinity pattern changes. We find that for time-periods at and less than 30 years, salinity pattern amplifications are highly volatile, with minimal amplification over the last 30 years, but large amplifications during the 1968-2001 time period. We conclude that time-periods greater than 40-years are needed if salinity pattern amplification is to be used as a proxy for changes in the hydrological cycle. Finally, we find no correlation between E-P and salinity pattern amplifications over the past 30-years.