Cyanobacteria, also known as blue-green algae, are microorganisms that can be found in fresh waterbodies, coastal waters, and offshore waters worldwide. They occur naturally, feeding themselves through photosynthesis and playing important roles in cycling nutrients and supporting the food web. Under certain environmental conditions, however, cyanobacteria accumulate and form algal blooms that can appear as a visible bluish-green scum. These blooms are unsightly and may have an unpleasant odor. More importantly, many species of cyanobacteria produce toxins (called cyanotoxins). Dense blooms can kill fish or other animals or harm people who are exposed to contaminated water through recreation or drinking water. Conventional drinking water treatment techniques can remove some cyanotoxins but may not be able to fully remove all the toxins during a bloom (U.S. EPA, 2018). A few recent large cyanobacterial blooms in U.S. lakes have caused drinking water utilities and state agencies to issue “do not drink” advisories. Cyanobacterial blooms are often driven by excess nutrients in the water—particularly phosphorus and nitrogen—as well as warmer water temperatures.

Scientists have measured cyanobacteria for many years by directly sampling selected drinking water reservoirs and recreational lakes. More recently, satellites have been used to estimate cyanobacteria concentrations in water, based on how blooms look. These satellite sensors are fundamentally detecting a combination of insignificant fluorescence and strong chlorophyll absorption from cyanobacteria. Satellite measurements complement onsite monitoring and can provide wider spatial and temporal coverage of cyanobacteria blooms.

This indicator uses satellite data to describe trends in detectable cyanobacteria in more than 2,000 lakes and reservoirs (ranging in size from about 0.75 km² to more than 4,000 km²) across the contiguous U.S. from 2008 to 2011 and 2017 to 2021. Detectable cyanobacteria have an abundance >20,000 cells per milliliter (see technical documentation). The indicator uses data from two European Space Agency satellite sensors to calculate cyanobacterial occurrence, area covered, and frequency. Satellite technology, however, cannot detect cyanotoxin levels or whether a given bloom is harmful. This indicator is based on standardized methods developed by the Cyanobacteria Assessment Network (CyAN), a collaboration among EPA, the National Aeronautics and Space Administration (NASA), the National Oceanic and Atmospheric Administration (NOAA), and the U.S. Geological Survey (USGS).

This indicator presents three related metrics focusing on national and regional-level data:

• Occurrence: the weekly percentage and number of lakes experiencing detectable cyanobacterial blooms (Exhibit 1).
• Area covered: the monthly surface area of detectable cyanobacterial blooms (Exhibit 2).
• Frequency: the percentage of weeks during the year that had a detectable bloom (Exhibit 3).

Additional data at the state and lake scale are available on EPA’s EnviroAtlas. For Exhibit 1, occurrence is defined as having detectable cyanobacteria in more than 10 percent of each lake’s surface area per week. The 10 percent threshold has been shown to reliably classify cyanobacterial occurrence.

Exhibit 1 summarizes the percentage and number of lakes and reservoirs across the contiguous U.S. and regions with detectable cyanobacteria. Cyanobacterial occurrence trends for the contiguous U.S. are consistent across years. At the national scale, the data show a recurring seasonal pattern with initiation of blooms around April (approximately 10 percent of lakes for 2008-2011 and 10 to 15 percent of lakes for 2017-2021), a maximum between September and November (25 to 30 percent of lakes for 2008-2011 and 30 to 40 percent of lakes for 2017-2021), and cyanobacteria going dormant in November and December, although wintertime imagery can be difficult to obtain due to snow and ice cover. In the northern regions, wintertime data were consistently limited due to snow and ice cover. Except in the South and Southeast, regional patterns follow the national seasonal pattern. In the South and Southeast, cyanobacterial blooms reach a maximum later in the year.

Exhibit 2 summarizes the monthly surface area (or spatial extent) of detectable cyanobacteria during the same timeframe. Spatial extent for the contiguous U.S. shows a consistent seasonal pattern across years, with maximum cyanobacterial extent (about 34,000 to 42,000 km² for 2008-2011 and about 40,000 to 52,000 km² for 2017-2021) mainly occurring between September and November. Across the contiguous U.S., cyanobacterial spatial extent has increased in recent years, driven primarily by increases in southern and northeastern states.

Exhibit 3 summarizes how frequently cyanobacterial blooms happen by showing the percentage of weeks in a year that cyanobacteria are detectable in an average lake or reservoir. Frequency trends for the contiguous U.S. and regions vary across years for each satellite sensor. No clear upward or downward trend is apparent. Across the contiguous U.S., the data show detectable cyanobacteria for about 18 percent of the year for 2008-2011 and 22 percent of the year for 2017-2021.

• This indicator only includes relatively large lakes that range in surface area from 0.75 to more than 4,000 km² to accommodate the spatial resolution of the satellite sensor (Urquhart and Schaeffer, 2020).
• The Great Lakes are not included in this dataset because they are monitored through a different program with different requirements, as prescribed by the Harmful Algal Bloom and Hypoxia Research and Control Amendments Act of 2014: https://www.govinfo.gov/content/pkg/BILLS-113s1254enr/pdf/BILLS-113s1254enr.pdf (PDF) (9 pp, 244k).
• Pixels along the edge of the lake had to be removed from the analysis because the data were part land, part water. This prevents the indicator from considering the shoreline, where blooms and recreation often occur. Higher-resolution sensors might provide more complete data (Coffer et al., 2020).
• Snow and ice interfere with these sensors’ ability to detect cyanobacterial blooms, so the data for cold regions can be limited to unfrozen portions of the year (Coffer et al., 2020).
• Satellite sensors cannot detect cyanobacteria in lower portions of the water column. In addition, wind stress can affect the vertical distribution of cyanobacteria throughout the water column (Coffer et al., 2020). Thus, this indicator is limited to blooms near the surface.
• This indicator has a 5-year gap due to a break in availability of satellite sensor data. The data before and after the gap come from two different satellite instruments and should not be compared directly with each other. Ongoing data collection should augment the existing record and eventually provide enough years of comparable data for identifying long-term trends in bloom characteristics.
• The land mask used during satellite processing filters out most perpetually dry lakes, but if these dry lakes fill at some future point in time, they will not be included in the dataset unless the land mask is updated.
• Satellite data processing does not account for changes in water levels, which can influence cyanobacterial bloom conditions, due to cycles such as drought and flood.

Data were obtained from CyAN, a collaboration of U.S. federal agencies using satellite measurements from two European Space Agency sensors: the MEdium Resolution Imaging Spectrometer (MERIS, 2008–2011) and the Ocean and Land Colour Instrument (OLCI, 2017 to present).