In May 2015, more than 35 trillion gallons of water fell over Texas, enough to cover the entire state 8 inches deep in water. Of course, this severe flooding had an impact on land; 11 people died and property were lost. But this significant amount of water had to have an impact on the ocean too.
NASA's Soil Moisture Active Passive (SMAP) satellite and other satellite instruments along with in situ data were used to create a comprehensive chronology of the flood from land to ocean. Besides SMAP Sea Surface Salinity (SSS) and soil moisture, the study used a wide array of other NASA observations: precipitation data from the Tropical Rainfall Measuring Mission (TRMM) and Global Precipitation Measurement (GPM) missions, water storage observations from the Gravity Recovery And Climate Experiment (GRACE) mission, ocean color observations from the Moderate Resolution Imaging Spectrometer (MODIS) instrument on board the Aqua satellite, and altimetric currents from the Jason satellite series.
The record rainfall in May 2015 produced saturated surface soil, limited infiltration of subsurface soil, and record river discharge. The 2010-2015 monthly time series represented in Figure 1 show that the May 2015 precipitation anomaly over Texas is the strongest during the entire record of the study (Figure 1a). It is associated with a saturation of surface soil moisture detected by SMAP, the strongest during the entire record of the study (Figure 1b) and with a record-length peak of river discharge over 6 years (Figure 1d). No record-length maximum is seen in terrestrial water storage after the flood but there is a local maximum relative to the ongoing drought (Figure 1c).
Once discharged into the ocean, the significant amount of freshwater loaded with sediments interacted with the regional circulation in the Gulf of Mexico and was traced by SSS and ocean color satellite data (Figure 2). The freshwater plume emerged from the Texas shelf in May 2015 and was carried along the coast in the northern part of the Gulf of Mexico in June and July. From July 2015, the plume was carried southward in the central gulf while the typical freshwater plume emerged from the Mississippi River mouth in the eastern part of the Gulf of Mexico. Together with the typical Mississippi River plume, the Texas freshwater plume caused an unusual freshwater plume forming a “horseshoe” from August to September 2015. The signature of the plumes in 2015 was also captured by the ESA Surface Moisture and Ocean Salinity (SMOS) mission. This unusual horseshoe pattern was not seen in 6 years of SMOS SSS observations or documented by in-situ observations. The riverine origin of this freshwater plume in the gulf in summer 2015 is confirmed by MODIS ocean color data measuring the light absorption in the ocean, another proxy for riverine waters (Figure 2, bottom panel). The transport of the Texas freshwater plume is confirmed by altimetric currents. During spring/summer 2015, there was an unusual strong and persistent Loop Current carrying the Mississippi River plume southeastward (Figure 3). An anticyclonic eddy was shed by the Loop Current in spring 2015 and persisted until the end of summer, carrying the freshwater plume along the northern coast of the gulf and then southward into the central gulf.
The rare occurrence of the horseshoe-shaped freshwater plume was caused by the freshwater plume from the Texas flood, the typical Mississippi River plume, an unusually strong Loop Current and its anticyclonic eddy to the west.
This particularly significant amount of terrestrial waters discharged and transported into the Gulf of Mexico can have potential biogeochemical implications by perturbing the ambient salinity and nutrients. The Gulf of Mexico is the largest hypoxic zone affecting the US (zone where not enough oxygen is available to longer support living aquatic organisms). Higher discharges tend to produce larger hypoxia zone, but also depend on other factors such as surface wind, currents, and vertical stratification.
PO.DAAC Science Team in collaboration with Severine Fournier