Chengfeng Le

A hybrid approach to estimate chromophoric dissolved organic matter in turbid estuaries from satellite measurements: A case study for Tampa Bay

Remote sensing of chromophoric dissolved organic matter (CDOM) from satellite measurements for estuaries has been problematic due to optical complexity of estuarine waters and uncertainties in satellite-derived remote sensing reflectance (Rrs, sr(-1)). Here we demonstrate a hybrid approach to combine empirical and semi-analytical algorithms to derive CDOM absorption coefficient at 443 nm (a(g)(443), m(-1)) in a turbid estuary (Tampa Bay) from MODIS Aqua (MODISA) and SeaWiFS measurements. The approach first used a validated empirical algorithm and a modified quasi-analytical algorithm (QAA) to derive chlorophyll-a concentration (Chla, mg m(-3)) and particulate backscattering coefficient at 443 nm (b(bp)(443), m(-1)), respectively, from which phytoplankton pigment and non-algal particulate absorption coefficient at 443 nm (a(ph)(443) and a(d)(443), m(-1)) were derived with pre-determined bio-optical relationships. Then, the modified QAA was used to estimate the total absorption coefficient at 443 nm (a(t)(443), m(-1)). Finally, a(g)(443) was estimated as (a(t)(443) - a(ph)(443) - a(d)(443) - a(w)(443)) where a(w)(443) is the absorption coefficient of pure water (a constant). Using data collected from 71 field stations and 33 near-concurrent satellite-field matchup data pairs covering a large dynamic range (0.3 - 8 m(-1)), the approach showed ~23% RMS uncertainties in retrieving a(g)(443) when in situ Rrs data (N = 71) were used. The same approach applied to satellite Rrs yielded much higher uncertainties of a(g)(443) (~85%) due to large errors in the satellite-retrieved Rrs(443). When the Rrs(443) was derived from the satellite-retrieved Rrs(550) and then used in the hybrid approach, uncertainties in the retrieved a(g)(443) reduced to ~30% (N = 33). Application of the approach to MODISA and SeaWiFS data led to a 15-year time series of monthly mean a(g)(443) distributions in Tampa Bay between 1998 and 2012. This time series showed significant seasonal and annual variations regulated mainly by river discharge. Testing of the approach over another turbid estuary (Chesapeake Bay, the largest estuary in the U.S.) demonstrated the potential (~25% uncertainties for a limited a(g)(443) range) of using this approach to establish long-term environmental data records (EDRs) of CDOM distributions in other estuaries with similar optical complexity.

Ocean Color Continuity From VIIRS Measurements Over Tampa Bay

Ocean color continuity calls for consistent observations from multiple sensors in order to establish a seamless data record to address earth science questions. Currently, both Moderate Resolution Imaging Spectroradiometer (MODIS) instruments on the Terra and Aqua satellites are being operated well beyond their designed five-year mission life, and they have shown signs of sensor degradation. It is thus urgent to evaluate whether the most recently launched Visible Infrared Imager Radiometer Suite (VIIRS) instrument (2011 to present) can provide consistent observations should MODIS instruments stop functioning. In this study, the consistency between MODIS/Aqua and VIIRS measurements over the Tampa Bay estuary ( ~ 1000 km2) is assessed for remote sensing reflectance (Rrs, sr-1), chlorophyll-a concentrations (Chla, mg·m-3), and absorption coefficient of colored dissolved organic matter (ag(443), m-1). While Rrs was derived as a standard National Aeronautics and Space Administration product from the SeaDAS software package (reprocessing version R2013.0), Chla and ag(443) were estimated using the recently developed regional algorithms for Tampa Bay. Time-series analysis and statistics both showed that the two sensors provided consistent measurements for most products evaluated, with unbiased mean percentage differences of 25% and mean annual biases within -9% (except for one of the eight cases) for large dynamic ranges in Chla (1.0-20 mg·m-3) and ag(443) (0.1-1.5 m-1) in all four bay segments. These estimates are comparable or better than those derived from satellite-in situ comparisons, suggesting that VIIRS will provide observations consistent with MODIS, ensuring ocean color continuity and seamless data records for Tampa Bay. Such observations are crucial in establishing a long-term satellite-based water quality decision matrix for Tampa Bay.

VIIRS Observations of a Karenia brevis Bloom in the Northeastern Gulf of Mexico in the Absence of a Fluorescence Band

The Visible Infrared Imager Radiometer Suite (VIIRS) is not equipped with a fluorescence band, which may affect its ability to detect and quantify harmful algal blooms (HABs) in coastal waters rich in colored dissolved organic matter. Such a deficiency has previously been demonstrated for a bloom of the toxic dinoflagellate Karenia brevis in the northeastern Gulf of Mexico (NEGOM) in summer 2014. Here, using data collected in the field and by VIIRS and Moderate Resolution Imaging Spectroradiometer (MODIS), we show that such a deficiency may be partially overcome using a red-green-chlorophyll-a index (RGCI). A relationship between near-concurrent (±4 hours) VIIRS RGCI (Rrs(672)/Rrs(551)) and field-measured chlorophyll-a (Chla; in mg m-3) was developed and evaluated using calibrated Chla obtained by a flowthrough system. A mean relative uncertainty, which was approximately twofold lower than VIIRS OC3M Chla, was obtained for VIIRS RGCI Chla (mean relative error: ~56%) over a large range (0.5-20 mg m -3). Similar spatial patterns between near-concurrent MODIS-Aqua (MODISA) normalized fluorescence line height (nFLH) and VIIRS RGCI Chla imagery indicate that VIIRS RGCI may be used as a surrogate for MODISA nFLH in the absence of a fluorescence band. The success of this newly developed data product may be partially attributed to the 20-nm bandwidth of the VIIRS 672-nm band (662-682 nm) that covers a portion of the solar stimulated fluorescence region. However, whether such observations from a simple case study can be extended to other turbid coastal or inland waters still remains to be tested.

Spatiotemporal chlorophyll-a dynamics on the Louisiana continental shelf derived from a dual satellite imagery algorithm

A monthly time series of remotely sensed chlorophyll-a (Chlars) over the Louisiana continental shelf (LCS) was developed and examined for its relationship to river discharge, nitrate concentration, total phosphorus concentration, photosynthetically available radiation (PAR), wind speed, and interannual variation in hypoxic area size. A new algorithm for Chlars, tuned separately for clear and turbid waters, was developed using field-observed chlorophyll-a (Chlaobs) collected during 12 cruises from 2002 to 2007. The new algorithm reproduced Chlaobs, with ∼40% and ∼60% uncertainties at satellite pixel level for clear offshore waters and turbid nearshore waters, respectively. The algorithm was then applied to SeaWiFS and MODIS images to calculate long-term (1998–2013) monthly mean Chlars estimates at 1 km resolution across the LCS. Correlation and multiple stepwise regression analyses were used to relate the Chlars estimates to key environmental drivers expected to influence phytoplankton variability. The Chlars time series covaried with river discharge and nutrient concentration, PAR, and wind speed, and there were spatial differences in how these environmental drivers influenced Chlars. The main axis of spatial variability occurred in a cross-shelf direction with highest Chlars observed on the inner shelf. Both inner (<10 m depth) and middle-shelf (10–50 m depth) Chlars were observed to covary with interannual variations in the size of the hypoxic (O2 < 63 mmol m−3) area, and they explained ∼70 and ∼50% variability in interannual hypoxia size, respectively.

Satellite‐based empirical models linking river plume dynamics with hypoxic area and volume

Satellite-based empirical models explaining hypoxic area and volume variation were developed for the seasonally hypoxic (O2 < 2 mg L−1) northern Gulf of Mexico adjacent to the Mississippi River. Annual variations in midsummer hypoxic area and volume were related to Moderate Resolution Imaging Spectroradiometer-derived monthly estimates of river plume area (km2) and average, inner shelf chlorophyll a concentration (Chl a, mg m−3). River plume area in June was negatively related with midsummer hypoxic area (km2) and volume (km3), while July inner shelf Chl a was positively related to hypoxic area and volume. Multiple regression models using river plume area and Chl a as independent variables accounted for most of the variability in hypoxic area (R2 = 0.92) or volume (R2 = 0.89). These models explain more variation in hypoxic area than models using Mississippi River nutrient loads as independent variables. The results here also support a hypothesis that confinement of the river plume to the inner shelf is an important mechanism controlling hypoxia area and volume in this region.

Satellite observation of particulate organic carbon dynamics on the Louisiana continental shelf

Particulate organic carbon (POC) plays an important role in coastal carbon cycling and the formation of hypoxia. Yet, coastal POC dynamics are often poorly understood due a lack of POC observations and the complexity of coastal hydrodynamic and biogeochemical processes that influence POC sources and sinks. Using a dataset of field observations and satellite ocean color products, we developed a new multiple regression algorithm to derive POC from satellite observations in two river-dominated estuaries in the northern Gulf of Mexico: the Louisiana Continental Shelf (LCS) and Mobile Bay. The algorithm had reliable performance with mean relative error (MRE) of ~40%, and root mean square error (RMSE) of ~50% for MODIS and SeaWiFS images in the two systems. Substantial spatio-temporal variability was observed from satellite on the LCS, with higher POC on the inner shelf (< 10 m depth) and lower POC on the middle (10-50 m depth) and outer shelves (50-200 m depth), and with higher POC in winter (January to March), and lower POC in summer to fall (August to October). Correlation analysis between long-term POC time series and several potential influencing factors indicated that river discharge dominants POC dynamics on the LCS. Wind and surface currents also affect POC spatial patterns on short time scales. This study demonstrates that algorithms that can determine coastal POC from satellites greatly increase the spatial and temporal extent of observations available for characterizing POC dynamics and their relations to various dominant physical forcings to the continental shelf and estuaries.

Satellite Derived Water Quality Observations Are Related to River Discharge and Nitrogen Loads in Pensacola Bay, Florida

Relationships between satellite-derived water quality variables and river discharges, concentrations and loads of nutrients, organic carbon, and sediments were investigated over a nine-year period (2003-2011) in Pensacola Bay, Florida, USA. These analyses were conducted to better understand which river forcing factors were the primary drivers of estuarine variability in several water quality variables. Remote sensing reflectance time-series data were retrieved from the MEdium Resolution Imaging Spectrometer (MERIS) and used to calculate monthly and annual estuarine time-series of chlorophyll a (Chla), colored dissolved organic matter (CDOM), and total suspended sediments (TSS). Monthly MERIS Chla varied from 2.0 mg m-3 in the lower region of the bay to 17.2 mg m-3 in the upper bay. MERIS CDOM and TSS exhibited similar patterns with ranges of 0.51 to 2.67 (m-1) and 0.11 to 8.9 (g m-3). Variations in the MERIS-derived monthly and annual Chla, CDOM, and TSS time-series were significantly related to monthly and annual river discharge and loads of nitrogen, organic carbon, and suspended sediments from the Escambia and Yellow rivers. Relationships differed, though, with monthly MERIS Chla most strongly correlated with river discharge lagged one-month, while monthly MERIS CDOM and TSS were most correlated with concurrent month river nitrate (NO3-) loads. Multiple regression models based on river loads (independent variables) and MERIS Chla, CDOM, or TSS (dependent variables) explained significant fractions of the variability (up to 62%) at monthly and annual scales. The most significant independent variables in the regressions were river nitrogen loads, which were associated with increased MERIS Chla, CDOM, and TSS concentrations, and river suspended sediment loads, which were associated with decreased concentrations. In contrast, MERIS water quality variations were not significantly related to river total phosphorus loads. The spatially synoptic, nine-year satellite record expanded upon the spatial extent of past field studies to reveal previously unseen system-wide responses to river discharge and loading variation. The results indicated that variations in Pensacola Bay Chla, CDOM, and TSS were primarily associated with riverine nitrogen loads. Thus, reducing these loads may improve water quality issues associated with eutrophication, turbidity, and water clarity in this system.