Description: Time series of in situ surface seawater partial pressure of CO2 (pCO2) data collected between 2005 and 2017, together with other environmental variables from field or satellite measurements, along the coasts of the United States of America and its territories at different latitudes, are analyzed to separate the temperature effect from the remaining non-temperature effects (i.e., biological and other physical effects) on driving surface pCO2. Similar to the findings in the open ocean, on seasonal time scales, the temperature effect (pCO2_T) tends to override the non-temperature effect (pCO2_nonT) in modulating surface pCO2 in tropical and subtropical oceanic waters. However, the balance between pCO2_T and pCO2_nonT tends to shift towards pCO2_nonT in temperate zone waters, with a few exceptions in some specific oceanic environments. On interannual time scales, both atmospheric pCO2 and surface pCO2 show significant increasing trends over short time scales (i.e., 〈10 years) except for a few outliers. In tropical and subtropical waters, the interannual changes of surface pCO2 are mainly controlled by the non-temperature effect (through air-sea CO2 exchange). In temperate regions, these changes are primarily driven by the temperature effect (through increased SST). Considering that temperature is commonly included in remote sensing algorithms of surface pCO2, this study suggests that, to better capture the seasonal and interannual signals in surface pCO2 from satellites, atmospheric pCO2 must be considered in the surface pCO2 remote sensing algorithms especially in tropical and subtropical waters. The non-temperature effect on surface pCO2, especially the biological effect (e.g., algal blooms), needs further investigation.

Description: “Whiting” events in the Bahama Banks, due to high concentrations of carbonate-rich particles suspended in the water, have been reported and discussed widely in the past 80 years. However, little is known about their distributions and particularly about their long-term changes. Here, using a deep learning (DL) model, we objectively delineate and quantify whiting features from Aqua MODIS (Moderate Resolution Imaging Spectroradiometer) satellite images (250-m resolution) and establish an 18-year data record (2003−2020) of whiting occurrences in the Bahama Banks. Both the Great Bahama Bank (GBB) and the Little Bahama Bank (LBB) show clear seasonality in whiting areas, where a primary peak in spring and a secondary peak in winter are found in the GBB but only one peak in winter is found in the LBB. Such a seasonality may be explained using a hydrodynamic hypothesis on calcium carbonate precipitation. The mean size of individual whiting patches in the GBB is about 2.4 ± 6.1 km2 (∼0.1 to 226 km2), while in the LBB is 1.4 ± 2.7 km2 (∼0.1 to 95 km2). The total whiting coverage in a typical cloud-free image is 87.1 km2 in the GBB and 32.0 km2 in the LBB, representing 0.14% and 0.76% of the entire GBB and LBB, respectively. Significant increases in the mean coverage have been found in the GBB since 2011, with peak coverage (∼200 km2) in 2013–2015 being at least 4 times higher than before (20–70 km2). Although the whiting area started to decrease after 2015, it did not reach the pre-2011 level until 2020. On the other hand, correlation analysis and principal component analysis of several environmental factors (pH, light, salinity, carbonate, aragonite, winds, currents) provided some hints on which factors may have contributed. From these, we infer a potential ‘Goldilocks’ scenario, whereby decreases in pH and carbonate concentration, concomitant with increases in Sea Surface Temperature (SST) and current speeds, created conditions increasingly favorable for whitings from 2011 to 2015. Continuation of these environmental trends after 2015, however, resulted in conditions increasingly unfavorable for whiting formation, yet without field-based measurements it is difficult to conclude the potential reasons for increases and decreases of whiting formation in the GBB.