Ryan Vandermeulen

Monitoring bio-optical processes using NPP-VIIRS and MODIS-Aqua ocean color products

Same day ocean color products from the S-NPP and MODIS provide for a new capability to monitor changes in the bio-optical processes occurring in coastal waters. The combined use of multiple looks per day from several sensors can be used to follow the water mass changes of bio-optical properties. Observing the dynamic changes in coastal waters in response to tides, re-suspension and river plume dispersion, requires sequential ocean products per day to resolve bio-optical processes. We examine how these changes in bio-optical properties can be monitored using the NPP and MODIS ocean color products. Additionally, when linked to ocean circulation, we examine the changes resulting from current advection compared to bio-optical processes. The inter-comparison of NPP and MODIS ocean products are in agreement so that diurnal changes surface bio-optical processes can be characterized.

Diurnal changes in ocean color in coastal waters

Coastal processes can change on hourly time scales in response to tides, winds and biological activity, which can influence the color of surface waters. These temporal and spatial ocean color changes require satellite validation for applications using bio-optical products to delineate diurnal processes. The diurnal color change and capability for satellite ocean color response were determined with in situ and satellite observations. Hourly variations in satellite ocean color are dependent on several properties which include: a) sensor characterization b) advection of water masses and c) diurnal response of biological and optical water properties. The in situ diurnal changes in ocean color in a dynamic turbid coastal region in the northern Gulf of Mexico were characterized using above water spectral radiometry from an AErosol RObotic NETwork (AERONET -WavCIS CSI-06) site that provides up to 8-10 observations per day (in 15-30 minute increments). These in situ diurnal changes were used to validate and quantify natural bio-optical fluctuations in satellite ocean color measurements. Satellite capability to detect changes in ocean color was characterized by using overlapping afternoon orbits of the VIIRS–NPP ocean color sensor within 100 minutes. Results show the capability of multiple satellite observations to monitor hourly color changes in dynamic coastal regions that are impacted by tides, re-suspension, and river plume dispersion. Hourly changes in satellite ocean color were validated with in situ observation on multiple occurrences during different times of the afternoon. Also, the spatial variability of VIIRS diurnal changes shows the occurrence and displacement of phytoplankton blooms and decay during the afternoon period. Results suggest that determining the temporal and spatial changes in a color / phytoplankton bloom from the morning to afternoon time period will require additional satellite coverage periods in the coastal zone.

Sensitivity of calibration gains to ocean color processing in coastal and open waters using ensembles members for NPP-VIIRS

The sensitivity of ocean color products to variations in vicarious calibration gains at Top of Atmosphere (TOA) shows varying impacts in different water types for Suomi- NPP VIIRS. Blue water vicarious gains from MOBY in situ data, which is used for global open waters, and green water gains derived from complex coastal WaveCIS AERONET waters, have a different impact on spectral normalized water leaving radiances and the derived ocean color products (inherent optical properties, chlorophyll). We evaluated the influence of gains from open and coastal waters by establishing a set of ensemble-processed products. The TOA gains show a non-linear impact on derived ocean color products, since gains affect multiple ocean color processing algorithms such as atmospheric correction, NIR iterations, etc. We show how the variations within the ensemble TOA gain members spatially impact derived products from different water types (high CDOM, high backscattering, etc). The difference in color products derived from the Blue and Green water gain show a spatial distribution to characterize the product uncertainty in coastal and open ocean water types. The results of the ensemble gain members are evaluated with in situ matchups. Results suggest the sensitivity of the ocean color processing for open ocean verses coastal waters.

Estimating Sea Surface Salinity in Coastal Waters of the Gulf of Mexico Using Visible Channels on SNPP-VIIRS

Sea surface salinity is determined using the visible channels from the Visual Infrared Imaging Radiometer Suite (VIIRS) to derive regional algorithms for the Gulf of Mexico by normalizing to seasonal river discharge. The dilution of river discharge with open ocean waters and the surface salinity is estimated by tracking the surface spectral signature. The water leaving radiances derived from atmospherically-corrected and calibrated 750-m resolution visible M-bands (410, 443, 486, 551, 671 nm) are applied to bio-optical algorithms and subsequent multivariate statistical methods to derive regional empirical relationships between satellite radiances and surface salinity measurements. Although radiance to salinity is linked to CDOM dilution, we explored alternative statistical relationships to account for starting conditions. In situ measurements are obtained from several moorings spread across the Mississippi Sound and Mobile Bay, with a salinity range of 0.1 - 33. Data were collected over all seasons in the year 2013 in order to assess inter-annual variability. The seasonal spectral signatures at the river mouth were used to track the fresh water end members and used to develop a seasonal slope and bias between salinity and radiance. Results show an increased spatial resolution for remote detection of coastal sea surface salinity from space, compared to the Aquarius Microwave salinity. Characterizing the coastal surface salinity has a significant impact on the physical circulation which affects the coastal ecosystems. Results identify locations and dissipation of the river plumes and can provide direct data for assimilation into physical circulation models.

Inter-satellite comparison and evaluation of Navy SNPP VIIRS and MODIS-Aqua ocean color properties

Navy operational ocean color products of inherent optical properties and radiances are evaluated for the Suomi–NPP VIIRS and MODIS-Aqua sensors. Statistical comparisons with shipboard measurements were determined in a wide variety of coastal, shelf and offshore locations in the Northern Gulf of Mexico during two cruises in 2013. Product consistency between MODIS-Aqua, nearing its end-of-life expectancy, and Suomi-NPP VIIRS is being evaluated for the Navy to retrieve accurate ocean color properties operationally from VIIRS in a variety of water types. Currently, the existence, accuracy and consistency of multiple ocean color sensors (VIIRS, MODIS-Aqua) provides multiple looks per day for monitoring the temporal and spatial variability of coastal waters. Consistent processing methods and algorithms are used in the Navy’s Automated Processing System (APS) for both sensors for this evaluation. The inherent optical properties from both sensors are derived using a coupled ocean-atmosphere NIR correction extending well into the bays and estuaries where high sediment and CDOM absorption dominate the optical signature. Coastal optical properties are more complex and vary from chlorophyll-dominated waters offshore. The in-water optical properties were derived using vicariously calibrated remote sensing reflectances and the Quasi Analytical Algorithm (QAA) to derive the Inherent Optical Properties (IOP’s). The Naval Research Laboratory (NRL) and the JPSS program have been actively engaged in calibration/validation activities for Visible Infrared Imager Radiometer Suite (VIIRS) ocean color products.

Diurnal changes in ocean color sensed in satellite imagery

Abstract. Measurements of diurnal changes in ocean color in turbid coastal regions in the Gulf of Mexico were characterized using above water spectral radiometry from a National Aeronautics and Space Administration (aerosol robotic network-WaveCIS CSI-06) site that can provide 8 to 10 observations per day. Satellite capability to detect diurnal changes in ocean color was characterized using hourly overlapping afternoon orbits of the visual infrared imaging radiometer suite (VIIRS) Suomi National Polar-orbiting Partnership ocean color sensor and validated with in situ observations. The monthly cycle of diurnal changes was investigated for different water masses using VIIRS overlaps. Results showed the capability of satellite observations to monitor hourly color changes in coastal regions that can be impacted by vertical movement of optical layers, in response to tides, resuspension, and river plume dispersion. The spatial variability of VIIRS diurnal changes showed the occurrence and displacement of phytoplankton blooming and decaying processes. The diurnal change in ocean color was above 20%, which represents a 30% change in chlorophyll-a. Seasonal changes in diurnal ocean color for different water masses suggest differences in summer and winter responses to surface processes. The diurnal changes observed using satellite ocean color can be used to define the following: surface processes associated with biological activity, vertical changes in optical depth, and advection of water masses.

Surface biomass flux across the coastal Mississippi shelf

The exchange of water masses across the Mississippi shelf was used to determine the chlorophyll flux for an eight month period in 2013 through the major Mississippi River discharge period in Spring and Fall. Circulation models (NCOM and HYCOM) and SNPP satellite chlorophyll products were used to monitor the changes in the shelf transport and surface biological impact. The physical and biological response of cross shelf exchange was observed in rapidly changing dynamic movements of river plumes across the shelf as identified by the models and satellite products. Six sections on the shelf identified exchange corridors of transport and biomass chlorophyll flux of surface waters between the coast and offshore waters. During the eight month period, the nearshore waters show high carbon chlorophyll flux, averaging -60 x103 kg chl extending to offshore waters. However, at the outer shelf break, a significant carbon flux was observed moving shoreward onto the shelf from offshore waters, averaging +100 x103 kg chl, which is attributed to the dynamic Mississippi River plume. Results indicate a significant amount of offshore surface waters containing biological carbon can exchange across the shelf, clearly demonstrated through the combination of biological satellite products and physical models.

Evaluation of VIIRS SST fields through the analysis of overlap regions between consecutive orbits

Full-swath Sea Surface Temperature (SST) derived from data acquired by the Visible Infrared Imaging Radiometer Suite (VIIRS) sensor on-board the Suomi-National Polar-orbiting Partnership (S-NPP) satellite produces significant overlap between consecutive orbits at all latitudes. In this study, we use those overlap regions to evaluate VIIRS SST, as inconsistencies between SST values from consecutive orbits are indications of likely degraded quality. The studies investigate two sources of inconsistencies: those resulting from the response of the SST equations when observing a scene from differing view angles and those caused by undetected data contamination. This study will present results for two VIIRS SST products: one from the Naval Oceanographic Office (NAVOCEANO), which is assimilated in the Navy Ocean Models, and the Advanced Clear-Sky Processor for Oceans (ACSPO) product from the National Oceanic and Atmospheric Administration (NOAA) Center for Satellite Applications and Research (STAR). Global statistics based on drifting buoys for both NAVOCEANO and NOAA products complete the analysis.

Regional vicarious gain adjustment for coastal VIIRS products

As part of the Joint Polar Satellite System (JPSS) Ocean Cal/Val Team, Naval Research Lab - Stennis Space Center (NRL-SSC) has been working to facilitate calibration and validation of the Visible Infrared Imaging Radiometer Suite (VIIRS) ocean color products. By relaxing the constraints of the NASA Ocean Biology Processing Group (OBPG) methodology for vicarious calibration of ocean color satellites and utilizing the Aerosol Robotic Network Ocean Color (AERONET-OC) system to provide in situ data, we investigated differences between remotely sensed water leaving radiance and the expected in situ response in coastal areas and compare the results to traditional Marine Optical Buoy (MOBY) calibration/validation activities. An evaluation of the Suomi National Polar-Orbiting Partnership (SNPP)-VIIRS ocean color products was performed in coastal waters using the time series data obtained from the Northern Gulf of Mexico AERONET-OC site, WaveCIS. The coastal site provides different water types with varying complexity of CDOM, sedimentary, and chlorophyll components. Time series data sets were used to develop a vicarious gain adjustment (VGA) at this site, which provides a regional top of the atmospheric (TOA) spectral offset to compare the standard MOBY spectral calibration gain in open ocean waters.

Comparison of VIIRS SST fields obtained from differing SST equations applied to a region covering the northern Gulf of Mexico and western North Atlantic

Several groups produce Sea Surface Temperature (SST) retrievals derived from data acquired by the Visible Infrared Imaging Radiometer Suite (VIIRS) sensor on-board the S-NPP satellite. Because of varying requirements or history, the groups often use differing SST equations to make their SST retrievals. Here we compare and discuss the equations through an examination of the SST fields. In most cases, the fields are created using the same program but differing equations, while in other cases, such as for the Interface Data Processing Segment (IDPS) Environmental Data Records (EDR), the SST fields are directly produced by other groups. Also discussed is the effect of the equation coefficients because independent groups may use the same equation but with different coefficients The focus of this study is on a region covering the Northern Gulf of Mexico and part of the Western North Atlantic. The comparison to buoys tries to minimize the effect of data contamination such as clouds on the results by matching the best satellite derived SST value in a neighborhood to the value from drifting or moored buoys. Finally we look at the overlap between consecutive passes to evaluate how the various equations perform at higher satellite zenith angles.