1. Learn
  2. Data Pathfinders
  3. Water Quality Data Pathfinder

Water Quality Data Pathfinder

New to using NASA Earth data? This pathfinder is designed to help guide you through the process of selecting and using applicable datasets, with guidance on resolutions and direct links to the data sources.

After getting started here, there are numerous NASA resources that can help develop your skills further. If you are new to remote sensing, check out What is Remote Sensing? or view the Applied Remote Sensing Training on Fundamentals of Remote Sensing.

When Hurricane Dorian passed over the Bahamas and along the southeastern United States coastline, its waves resuspended large quantities of sea-floor sediment which give the ocean a milky, aquamarine appearance in the above composite of VIIRS data collected on September 7, 2019. The browner hues closer to the U.S. shore come from runoff generated by the heavy rainfall of the hurricane.

When Hurricane Dorian passed over the Bahamas and along the southeastern United States coastline, its waves resuspended large quantities of sea-floor sediment which give the ocean a milky, aquamarine appearance in the above composite of VIIRS data collected on September 7, 2019.

Water quality managers and researchers have a critical need to monitor bodies of water locally, regionally, and globally to determine impacts to ecosystems, humans, and the environment and to restore and protect coastal and surface waters. A combination of ground- and satellite-based tools provides a more regional to global understanding of the impacts of water quality. These measurements help scientists, researchers, and decision makers to forecast events and assess conditions in near real-time so as to make timely decisions.

About Ocean Color Data

About Ocean Color Data

NASA, in collaboration with other organizations, has various instruments that provide information for understanding a number of phenomena associated with water quality. NASA’s Earth science data products are validated, meaning the accuracy has been assessed over a widely distributed set of locations and time periods via several ground-truth and validation efforts.

Satellites and sensors referenced in this pathfinder include:

Satellite

Sensor

Spatial Resolution

Temporal Resolution

Landsat 7

Enhanced Thematic Mapper (ETM)

15, 30, 60 m

16 days

Landsat 8

Operational Land Imager (OLI)

15, 30, 60 m

16 days

Terra and Aqua

Moderate Resolution Imaging Spectroradiometer (MODIS)

250 m, 500 m, 1 km

1-2 days

Suomi NPP

Visible Infrared Imaging Radiometer Suite (VIIRS)

375-750 m

1-2 days

Sentinel 2

Multi Spectral Imager (MSI)

10, 20, 60 m

5 days

Sentinel 3

Ocean and Land Color Instrument (OLCI)

300 m

2 days

Benefits and Limitations of Ocean Color Data

Benefits and Limitations of Ocean Color Data

The United States is fortunate to have numerous in-situ measurements for assessing water quality parameters, yet in-situ measurements have limited sample collection and so are not representative of the entire water body. In other countries and in more rural areas of the United States, sampling is even more limited or non-existent. Satellite data provide more regional to global spatial coverage; some information is available in near real-time, allowing for a more efficient response. Satellite data have also been collected for a longer period of time, providing for data continuity and trend analyses. With satellite data, assessments can be made regarding ocean color, but this provides only qualitative measures. For quantitative water quality monitoring analysis, in-situ measurements are required; the combination of satellite observations with in-situ makes for a more robust and integrated forecasting and response system.

While satellite data provide a more global view, it is important to note that satellite measurements are made through the atmosphere and not at the water level. As such, atmospheric correction algorithms must be run before water quality assessments can be made.

Also note that the sensors all have varying spatial, temporal and radiometric resolutions (for more detailed information on resolutions, review What is Remote Sensing). For example, many of the polar-orbiting satellites only pass over the same location every 1-2 days, but have a coarser spatial resolution, while others pass over every 16+ days, but have a much finer spatial resolution. Finding the right instrument or understanding the modeling processes for your area of interest is key.

Other challenges include the difficulty in separating water quality parameters of colored dissolved organic matter (CDOM), non-algal properties (NAP), and chlorophyll content when all three are present. Also, remote sensing observations alone are unable to discern between algal types or toxins.

Tools for Data Access and Visualization

Tools for Data Access and Visualization

Earthdata Search provides a means of searching all of NASA’s Earth science data across all Distributed Active Archive Centers (DAACs). It provides the only means to search all data regardless of where the data are archived. Within Earthdata Search, you can subset using temporal and geographic constraints. Some data can be customized after the data of interest are selected; to do this, add the desired data to your project and then click Download All.

Screenshot of the Search Earthdata site.

In the project area, you can select to customize your granule. You can reformat the data and output as HDF, NetCDF, ASCII, KML or a GeoTIFF. You can also choose from a variety of projection options. Lastly you can subset the data, obtaining only the bands that are needed.

Earthdata Search customization tools diagram.

Panoply

HDF and NetCDF files can be viewed in Panoply, a cross-platform application that plots geo-referenced and other arrays. Panoply offers additional functionality, such as slicing and plotting arrays, combining arrays, and exporting plots and animations.

NASA's National Snow and Ice Data Center Distributed Active Archive Center (NSIDC DAAC) has an HDF to GeoTIFF conversion tool, which allows you to geolocate, subset, stitch, and regrid certain HDF-EOS datasets.

Giovanni

Giovanni is an online environment for the display and analysis of geophysical parameters. There are a few options for analysis.

  • Time-averaged maps are a simple way to observe the variability of data values over a region of interest.
  • Map animations are a means to observe spatial patterns and detect unusual events over time.
  • Area-averaged time series are used to display the value of a data variable that has been averaged from all the data values acquired for a selected region for each time step.
  • Histogram plots are used to display the distribution of values of a data variable in a selected region and time interval.

For more detailed tutorials:

  • Giovanni How-To’s on the NASA GES DISC YouTube channel.
  • Data recipe for downloading a Giovanni map, as NetCDF, and converting its data to quantifiable map data in the form of latitude-longitude-data value ASCII text.

Worldview

Worldview's Events tab provides information about events, such as tropical cyclones, wildfires, volcanic eruptions, and even large iceberg movement. Hurricane Barry, as shown in this image, traveled from the Gulf into Louisiana in July 2019.
Worldview is an interactive interface for browsing full-resolution, global satellite imagery.
Worldview is a visualization tool to interactively browse global, full-resolution satellite imagery layers and then download the underlying data. Many of the imagery layers are updated within three hours of observation, essentially showing the entire Earth as it looks "right now." View current natural hazards and events using the Events tab which reveals a list of natural events, including wildfires, tropical storms, and volcanic eruptions. You can animate the imagery over time or do a screen-by-screen comparison of data for different time periods or a comparison of different datasets.

Worldview data visualization of the nighttime lights in Puerto Rico pre- and post- Hurricane Maria, which made landfall on September 20, 2017. Post-hurricane image shows widespread outages around San Juan, including key hospital and transportation infrastructure.

Worldview data visualization of the nighttime lights in Puerto Rico pre- and post- Hurricane Maria, which made landfall on September 20, 2017. Post-hurricane image shows widespread outages around San Juan, including key hospital and transportation infrastructure.

SeaDAS

SeaDAS is a comprehensive software package for the processing, display, analysis, and quality control of ocean color data. While the primary focus of SeaDAS is ocean color data, it is applicable to many satellite-based earth science data analyses.

SeaDAS is a comprehensive software package for the processing, display, analysis, and quality control of ocean color data. This image shows ocean color, sea surface temperature and non-algal material plus colored dissolved organic matter.

SeaDAS is a comprehensive software package for the processing, display, analysis, and quality control of ocean color data. This image shows ocean color, sea surface temperature and non-algal material plus colored dissolved organic matter.

SeaDAS processing components (OCSSW)

SeaDAS processing components (OCSSW)

Within SeaDAS, you can visualize data, and re-project, crop and create land, water and coastline masks. You can perform mathematical and statistical operations, such as band math (band additions and subtractions, band ratios), plot histograms, scatter plots and correlation plots, and you can incorporate in-situ data.

If you only have reflectance values, you can use the band ratio and algorithm coefficients within SeaDAS to derive chlorophyll-a. If using Landsat data, you need to convert Level 1 to Level 2 data. To do this, make sure your data processors within SeaDAS are updated.

In-situ data can be incorporated as well; this is critical for data validation. To integrate in-situ data, whether from SeaBASS or from another source, the data must be in a specific format. The file must be tab-delimited with fields of data, time, station (with the stations defined in the file), lat, lon, depth. Date and time are relevant as well. They need to be defined as YYYYMMDD and time as HH:MM:SS. If not defined properly, the file must be reworked to make it readable.

Once the tab-delimited file is complete, you can select Vector/Import and then select your data source. Remember in order to validate your remotely sensed data, you only want to look at the in-situ data at the surface (depth of 0).

SeaDAS allows for the integration of in-situ data in order to validate satellite measurements.

SeaDAS allows for the integration of in-situ data in order to validate satellite measurements.

For more detailed tutorials:

  • SeaDAS Video tutorials and demos—OB.DAAC recommends viewing the first few in the order they are shown. The core videos are listed first, followed by multi-tool case studies; everything below that appears in chronological order by release date.
  • SeaDAS FAQs—Frequently asked questions from SeaDAS users.

Use Ocean Color Data

Use Ocean Color Data

Satellite imagery, coupled with in-situ data, aids in the assessment of water quality by distinguishing between parameters, such as dissolved organic matter, sediments, plankton, and algal blooms. Some of these parameters affect optical properties which allows for their detection by remote sensors. For example, colored dissolved organic matter (CDOM) is a mixture of organic substances produced as organic matter decays, such as tannins; tannins stain the water (making it look black) affecting light absorption.

Scientists, researchers, water resource managers, decision makers, and others use remotely sensed data in various ways, including to help meet the 17 sustainable development goals put forth by the United Nations. Specifically, goal 6 states that countries should ensure availability and sustainable management of water and sanitation for all. (To see the data in use, view our Data User Profiles or our Freshwater Feature Articles.)

Ocean Color

Absorption equation, the total of which is equal to the absorption of phytoplankton, non-algal properties, colored dissolved organic matter, and water.

Absorption equation, the total of which is equal to the absorption of phytoplankton, non-algal properties, colored dissolved organic matter, and water.

Ocean color remote sensing uses remote-sensing reflectance, which is based on the properties of the materials in the water. When light interacts with water, it can be absorbed or scattered. Light is absorbed by a combination of phytoplankton, non-algal properties (NAP), CDOM and water itself.

Through a series of complex algorithms, the relationship between this absorption and scattering (in forward and backward directions), can provide a remote-sensing reflectance value for the water-leaving radiance.

When light interacts with water, it can be absorbed or scattered. Through a series of complex algorithms, the relationship between this absorption and scattering can provide a remote-sensing reflectance value for the water-leaving radiance.

When light interacts with water, it can be absorbed or scattered. Through a series of complex algorithms, the relationship between this absorption and scattering can provide a remote-sensing reflectance value for the water-leaving radiance.

One important note is that satellite sensors measure the top of atmosphere radiances. These radiances result from a combination of surface and atmospheric conditions (effects of clouds and aerosol particles); in order to look at just the water-leaving reflectance, the data must be atmospherically corrected; the goal is to subtract the atmospheric path and the surface-reflected spectrum and retain only the water-leaving radiance spectrum. There are a number of free products which help do this (see Find Ocean Color Data, below).

Once atmospheric correction has been applied, ocean color reflectance can be visualized and a qualitative interpretation made based on color. Generally, chlorophyll appears as green, water as blue, CDOM as black, and sediments/NAP as browns and whites. Every element has a spectral fingerprint, a unique absorption/reflectance pattern at varying wavelengths of the electromagnetic spectrum, which aids in the identification of these different parameters.

Spectral fingerprints, unique absorption/reflectance patterns at varying wavelengths of the electromagnetic spectrum, of ocean color parameters.

Spectral fingerprints, unique absorption/reflectance patterns at varying wavelengths of the electromagnetic spectrum, of ocean color parameters.

Ocean color reflectance can be visualized and a qualitative interpretation made based on color. Generally, chlorophyll appears as green, water as blue, CDOM as black, and sediments/NAP as browns and whites.

Ocean color reflectance can be visualized and a qualitative interpretation made based on color. Generally, chlorophyll appears as green, water as blue, CDOM as black, and sediments/NAP as browns and whites.

For a more quantitative measurement of ocean color data, different algorithms combine atmospherically corrected satellite images and in-situ measurements. Having in-situ data for validation is critical to getting a quantitative measure.

Find Ocean Color Data

Find Ocean Color Data

This May 10, 2019 Aqua/MODIS image exhibits patterns driven by the same physics as the patterns found in the Benguela upwelling a year earlier. Recent measurements in part of the region indicate that the ocean here is a source of the potent greenhouse gas, nitrous oxide.

This May 10, 2019 Aqua/MODIS image exhibits patterns driven by the same physics as the patterns found in the Benguela upwelling a year earlier. Recent measurements in part of the region indicate that the ocean here is a source of the potent greenhouse gas, nitrous oxide. https://oceancolor.gsfc.nasa.gov/gallery/635/

Ocean color is measured based on the amount of absorption by particles, (e.g., phytoplankton, sediments, colored dissolved organic matter (CDOM)) and in turn, the amount of water-leaving radiance. Having a quantitative measure of these parameters is useful in understanding how water bodies, such as the ocean, are evolving, as well as determining the quality of the water for consumption by living organisms. The primary means of measuring ocean color from space is through Landsat, the Terra and Aqua satellites, the joint NASA/NOAA Suomi National Polar-orbiting Partnership (Suomi NPP), and the European Space Agency’s suite of Sentinel missions. Each of these satellites has sensors acquiring data at different spatial, temporal, spectral, and radiometric resolutions (for detailed information on these, read What is Remote Sensing?).

In addition to ocean color, sea surface temperature (SST) is a valuable parameter as warmer waters can contribute to the growth of algal blooms. However, in the ocean cold upwelling waters usually bring nutrients from the seafloor fueling marine phytoplankton blooms. In the MODIS and VIIRS data, there is also an inherent optical properties (IOP) file, which provides an estimate of reflectance by CDOM. Specifically, the adg_443_giop is the absorption coefficient of non-algal material plus CDOM.

For more information on the algorithm used to generate this product and others, view the Ocean Biology DAAC (OB.DAAC) algorithm descriptions.

Science quality (higher-level “standard”) data products can be accessed via Earthdata Search or through NASA partner websites:

  • Landsat Data from USGS Earth Explorer
    Landsat is a joint NASA/USGS program that provides the longest continuous space-based record of Earth’s land in existence. On the Earth Explorer site, specify your search criteria, then:
    • select “Data Sets”
    • select Landsat
    • Select Landsat Collection 1 Level-1
    • Select Landsat 7 and/or Landsat 8

    These files can be downloaded as Level-1 GeoTIFF Data Products. Note that you will need a USGS login to proceed and that you will need to atmospherically correct the image. Acolite is one free tool that performs this correction. FLAASH is a function in the ENVI image processing program that also performs this correction. SeaDAS will convert Level 1 Landsat to Level 2 once atmospheric correction has been done.

Data products can be visualized as a time-averaged map, an animation, seasonal maps, scatter plots, or a time series through an online interactive tool, Giovanni. Follow these steps to plot data in Giovanni: 1) Select a map plot type. 2) Select a date range. Data are in multiple temporal resolutions, so be sure to note the start and end date to ensure you access the desired dataset. 3) Check the box of the variable in the left column that you would like to include and then plot the data. For more information on choosing a type of plot, see the Giovanni User Manual.

  • Level 3 data products from OB.DAAC
    Data products include chlorophyll concentration, SST, reflectance, and other related measurements from MODIS and VIIRS at 4 km and 9 km resolution. These data products are provided in five temporal resolutions: daily, 8-day, monthly, seasonally, and annually.
  • Aqua MODIS Chlorophyll-a Concentration data from Giovanni
    Data products from MODIS on the Aqua satellite at 4 km resolution provided at both 8-day and monthly temporal resolutions.
  • Aqua MODIS SST data from Giovanni
    Data products from MODIS on the Aqua satellite at 4 km resolution provided at both 8-day and monthly temporal resolutions.

Near real-time (NRT) data can be accessed via Worldview:

Ocean Color in Freshwater Inland Lakes

Ocean Color in Freshwater Inland Lakes

Monitoring ocean color in freshwater lakes is critical as assessing cyanobacteria, pathogens such as E.coli, man-made pollutants, nutrient inputs, and water clarity have implications for drinking water for humans and domestic animals, wildlife, and other ecosystems. Because these water bodies tend to be smaller and are surrounded by land, it is imperative to look at the sensor qualities before deciding which one to use.

This Sentinel 2A image of western Lake Superior was collected on May 7, 2019.

This Sentinel 2A image of western Lake Superior was collected on May 7, 2019.

The size of the water body must be considered. For large lakes like Lake Victoria or one of the Great Lakes, MODIS and VIIRS data are adequate. They provide a coarser spatial resolution but a more frequent temporal resolution, which is useful in a dynamic system. For small lakes, Landsat and Sentinel are needed as they provide the fine spatial resolution needed, but note that the temporal resolution is coarse, allowing for only monthly and seasonal monitoring.

In SeaDAS, converting the data of inland water bodies from Level 1 to Level 2 requires modification of the algorithm criteria. It is often more effective to turn off aerosol subtraction, as well as cloud, tilt, and land masks due to the algorithm resolution. For more information, check out the Applied Remote Sensing Training (ARSET) webinars, Integrating Remote Sensing into a Water Quality Monitoring Program, Part 2, and Remote Sensing for Freshwater Habitats, which focuses on inland water bodies and processing of related data.

In-situ Data

In-situ Data

The R/V Sally Ride, operated by the Scripps Institution of Oceanography, before departing for the northeastern Pacific Ocean to collect detailed ship-based measurements of plankton.

The R/V Sally Ride, operated by the Scripps Institution of Oceanography, before departing for the northeastern Pacific Ocean to collect detailed ship-based measurements of plankton.

In-situ data are easily integrated within SeaDAS and are available through the SeaWIFS Bio-optical Archive and Storage System (SeaBASS). SeaBASS contains in-situ measurements of apparent and inherent optical properties, phytoplankton pigment concentrations, and other oceanographic and atmospheric data.

If you are planning to collect your own in-situ data timed to when the satellite passes over your location, use the Overpass Prediction Tool.

The Bio-Optical in situ Data Discovery and Access with SeaBASS webinar covers how SeaBASS can be leveraged for data search, discovery, and access, and will demonstrate how SeaBASS supports NASA’s ocean color satellite products and the broader scientific community.

Other NASA Assets of Interest

Other NASA Assets of Interest

The Applied Remote Sensing Training program has numerous water quality training resources. There are resources on integrating remote sensing into water quality management, water quality in freshwater systems/inland bodies, harmful algal blooms (HAB), and addressing sustainable development goal 6.

For questions regarding ocean color data from NASA and processing or analysis of that data, the OB.DAAC has an Ocean Color User Forum.

Discover NASA Ocean Color Data, Services and Tools webinar introduces you to the science of ocean color, the activities and services of NASA's Ocean Biology Processing Group (OBPG), and shows you how to discover, access, utilize, and analyze NASA's ocean color data products.

The Plankton, Aerosol, Cloud, ocean Ecosystem (PACE) mission, scheduled to launch in 2022, contains an Ocean Color Instrument (OCI), which will measure properties of light at finer wavelength resolution than previous NASA satellite sensors, extending key system ocean color data records for climate studies.

External Resources

External Resources

The multi-agency (EPA, NASA, NOAA, and USGS) Cyanobacteria Assessment Network, or CyAN, has developed an app to alert officials and members of the public when a harmful algal bloom (HAB) could be forming, depending on specific changes in the color of the water observed by satellites. A NASA article on this effort is available as NASA Helps Warn of HABs.

  • Cyanobacteria Assessment Network, or CyAN logo

    Cyanobacteria Assessment Network, or CyAN logo

    Cyanobacteria Index from CyAN:
    Cyanobacteria HABs are a big problem in lake bodies. Not only do they produce excessive biomass, which consumes oxygen needed by other living organisms, but they also produce toxins. Many of these cyanobacteria species produce surface scum and so can be detected using remote sensing reflectance. Calculating the cyanobacteria index provides an indication of cyanobacteria bloom. The algorithm plots slope between 665-709 nm and then looks at the difference between the slope and the bloom spectral fingerprint. If the spectrum is below the slope line, it’s a negative, indicating a cyanobacteria bloom.

CyAN provides information on the cyanobacteria index.

The National Water Quality Monitoring Council Portal is a cooperative service sponsored by the United States Geological Survey (USGS), the Environmental Protection Agency (EPA), and the National Water Quality Monitoring Council (NWQMC). It serves in-situ data collected by over 400 state, federal, tribal, and local agencies.

The European Space Agency’s Sentinel Hub provides technical information on the missions, access to the data, user guides and tutorials, and news related to Sentinel data. In addition, their Water Quality Monitoring page provides information on their suite of products that can be used for monitoring.

NOAA’s Great Lakes Environmental Research Laboratory (GLERL) forecasts HABs within Lake Erie with their experimental HAB Tracker. They also have a GLERL YouTube Channel, which provides information on 2019 conditions and outlooks, harmful algal blooms including cyanobacteria, and climate-based projections.

NOAA’s National Centers for Coastal Ocean Science developed the Algal Bloom Monitoring System to routinely deliver near real-time products for use in locating, monitoring, and quantifying algal blooms in coastal and lake regions of the US.

UNESCO’s Water Quality Portal is a free visualizer of satellite-derived water quality information for worldwide lakes and rivers. A global set of parameters in 90 m spatial resolution is provided on a continental base.

Last Updated: Oct 28, 2019 at 8:36 AM EDT