update sentinel datasets summary and titles

datasets.d/development/sentinel_3A_POLYMER.xml

@@ -42,7 +42,7 @@
         <att name="geospatial_lon_min">-139</att>
         <att name="geospatial_vertical_min">0</att>
         <att name="geospatial_vertical_max">5</att>
-        <att name="institution">Spectral Labratory,Hakai Institute</att>
+        <att name="institution">Hakai Institute</att>
         <att name="project">Oceanography</att>
         <att name="product_version">1</att>
         <att name="license">https://creativecommons.org/licenses/by/4.0</att>
@@ -52,24 +52,20 @@
         <att name="keywords">remote sensing,chlorophyll a,phytoplankton,coastal zone,water quality,abundance and biomass</att>
         <att name="platform">satellite</att>
         <att name="platform_vocabulary">http://vocab.nerc.ac.uk/collection/L06/current/</att>
-        <att name="summary">This is an ongoing dataset of automated daily POLYMER processed Sentinel-3A OLCI images of ocean chlorophyll-a concentrations. Sentinel-3A is an oceanography focused European Space Agency (ESA) satellite that is jointly operated with the European Organization for the Exploitation of Meteorological Satellites (EUMSTAT). The system has been operational since April 2016 and was designed to continue the Envisat Medium Resolution Imaging Spectrometer (MERIS) mission. The OLCI is a push-broom radiometer that has a 300m spatial and daily temporal resolution, 21 spectral bands from 400-1200nm, an improved signal-to-noise ratio and an off-nadir swath that is centered to minimize ocean sun glint. These features make the instrument well suited to retrievals of biogeochemical products from optically complex coastal waters.
-
-            Satellite retrievals of biogeochemical products from optically complex coastal regions can have high error and as a result, the performance of OLCI is under continuous review. In British Columbia, the SPECTRAL remote sensing laboratory at the University of Victoria has performed extensive evaluation of atmospheric correction methods to determine the most precise approaches for regional chlorophyll retrievals. Validation of results with in-situ data showed the best results using OLCI Sentinel-3A Level-1 images processed with the POLYnomial based algorithm applied to MERIS (POLYMER) atmospheric correction with the application of the recommended flags including the &quot;Case-2&quot; flag and exclusion of the &quot;Thick Aerosol&quot; flag.  Detailed information on these validations and results are provided in Giannini et al. (2021) and POLYMER in Steinmetz et al. (2016) provided in the accompanying resources.
-
-            Following the derivation of the best approach, the SPECTRAL laboratory in collaboration with the University of Victoria MOD(ularity) Squad, developed an automated system that outputs near real-time (prior day) fully processed OLCI Sentinel 3A chlorophyl imagery for the British Columbia and southeast Alaska coast and offshore waters. In brief, the system: 1) downloads Sentinel-3A Level-1 data from the Sentinel-3 Marine Copernicus Online Data Access (CODA) web service; 2) applies the POLYMER atmospheric correction and flagging and; 3) mosaics the imagery for the study region resulting in fully processed level-3 chlorophyll imagery. Details on the development of this system are provided in Jacoby et al. (2019) referenced in the accompanying resources.
-
-            In 2022, the Hakai Institute took continued responsibility for the processing pipeline. Since then, the Hakai institute has transferred processing infrastructure from Microsoft Azure to the Amazon Web Service (AWS) and created automated submission of daily fully processed chlorophyll imagery to the Canadian Integrated Ocean Observing System (CIOOS). 
-
-            This product provides the current best known regional OLCI Sentinel-3A chlorophyll retrievals that were shown to have low systematic biases; however, relative percent differences remain high (80-100%) and this should be acknowledged by data users. In addition, error often exists in high turbidity regions (Fraser River plume and fjords) and at times, these pixels were difficult to correct or remove from the imagery and care should be taken when using data from these areas. Finally, there is continued evaluation and improvement of all aspects of the above methods and new methods may provide better retrievals than those provided here.
-
-
-            Satellite remote sensing has been increasingly used to study surface ocean processes over large areas and high temporal resolutions that would not be achievable using ship based methods. Chlorophyll, as the most widely used measure of phytoplankton biomass, is crucial for understanding phytoplankton which are the base of marine food webs and crucial for ocean biogeochemical cycling. Satellite retrievals of chlorophyll provide the necessary resolution required to understand long term variability under a changing climate.
-
-
-            Funding for the SPECTRAL laboratory validation studies and processing automation was provided by the UBC/UVic Hakai Coastal Initiative postdoctoral fellowship, NSERC NCE Marine Environmental Observation Prediction And Response (MEOPAR) network, Canadian Space Agency (CSA), Canadian Foundation for Innovation (CFI) and NSERC Discover Grant. 
+        <att name="summary">This is an ongoing dataset of fully processed daily Sentinel 3A and 3B chlorophyll-a (Chla) and suspended particulate matter (SPM) imagery for coastal and offshore British Columbia (BC) and Southeast Alaska waters. Setinel 3A and 3B are European Space Agency (ESA) oceanography satellites jointly operated with the European Organization for the Exploitation of Meteorological Satellites (EUMETSAT). The ocean and land colour instrument (OLCI) onboard both satellites has a 300m spatial resolution, near daily temporal coverage (when 3A and 3B are combined), 21 spectral bands from 400-1200nm, high signal-to-noise ratio and an off-nadir swath centered to minimize ocean sun glint. These features make the instruments well suited to retrievals of biogeochemical products from optically complex coastal waters.
 
-            The data originators (SPECTRAL remote sensing laboratory) request that users reference Giannini et al. (2021) and Jacoby et al. (2019) if data is used for published research.</att>
-        <att name="title">Sentinel-3A OLCI Imagery - Automated daily POLYMER processed satellite chlorophyll concentrations for coastal British Columbia and southeast Alaska</att>
+            At the University of Victoria (BC, Canada), the SPECTRAL remote sensing laboratory has performed extensive evaluation of methods for the best regional Chla and SPM retrievals. Validation with in-situ data showed the best results using Level-1 imagery processed with the POLYnomial based algorithm applied to MERIS (POLYMER) processor. Following validation, the SPECTRAL laboratory and the MOD(ularity) Squad developed an automated processing system that: 1) downloads imagery from the Marine Copernicus Online Data Access (CODA) web service; 2) applies POLYMER and flagging and; 3) mosaics the imagery for fully processed Chla and SPM concentrations over the study region. Additionally, an interactive public web interface was created to view the near real time outputs at www.algaeexplorer.ca (provided in resources). Full validation details are provided in Giannini et al. (2021) and processing details in Jacoby et al. (2019) and Marchese et al. (2022) referenced in the resources.
+
+            In 2022, the Hakai Institute took responsibility of the project, added processing of 3B imagery, created automated submission to the Canadian Integrated Ocean Observing System (CIOOS) and updated the Algae Explorer web interface.
+
+            This product provides the best known regional OLCI Chla and SPM retrievals shown to have low systematic biases (&lt;1%) and follow expected seasonal and spatial trends; however, relative percent difference between validation data and satellite retrievals was high notably for Chla (~83%) due to the underestimation of high Chla concentrations and potentially an artifact of spatial-temporal mismatches between validation samples and image pixels. The SPECTRAL laboratory has performed initial radiometric comparisons of 3A and 3B imagery and shown highly comparable data; however, comparison of biogeochemical outputs is still in progress. Further uncertainties exist in high turbidity regions (Fraser River plume and fjords) where uncorrectable poor-quality pixels are sometimes observed. Additionally, an unresolved data striping issue is periodically present and observed as a narrow band of distorted pixels, which sometimes evades the land mask and, crosses images diagonally above Vancouver Island. Efforts are underway to correct this issue. Data users should consider these uncertainties and issues when using the data. 
+
+            Satellite remote sensing is increasingly used to study surface ocean processes at the spatial and temporal resolutions required for understanding long term variability under a changing climate. Chlorophyll-a is the most widely used measure of phytoplankton biomass and crucial for understanding phytoplankton which are the base of the marine food web and control ocean biogeochemical cycling. Suspended particulate matter is a key water quality indicator (i.e. turbidity) with increased concentrations reducing light availability to aquatic species. 
+
+            Funding for validation and automation development was provided by the UBC/UVic Hakai Coastal Initiative postdoctoral fellowship, NSERC NCE Marine Environmental Observation Prediction and Response (MEOPAR) network, Canadian Space Agency (CSA), Canadian Foundation for Innovation (CFI) and NSERC Discover Grant awarded to Maycira Costa. 
+
+            It is requested that Giannini et al. (2021), Jacoby et al. (2019) and Marchese et al. (2022) are referenced if data is used for published research and the ESA acknowledged as the data provider.</att>
+            <att name="title">Daily satellite (Sentinel 3A and 3B) chlorophyll and suspended matter concentrations for coastal British Columbia and southeast Alaska</att>
     </addAttributes>
     <axisVariable>
         <sourceName>***pathName,timeFormat=yyyy/MM/dd,/algae_explorer/5_erddap_data/(\d{4}/\d{2}/\d{2})/polymer_mosaic_output\.nc,1</sourceName>